COLUMBIA LIBRARIES OFFSITE HEALTH SCIENCES STANDARD Hill "" HX64141810 Q P525 . B38 Colloids in biology ^ Columbia ®mbemtt> in t\}t Citp of i^ebJ ^or u ^tfjool of Bental anb (J^ral g^urgerp 3^eference l^ibrarp K Digitized by the Internet Archive in 2010 with funding from Open Knowledge Commons (for the Medical Heritage Library project) http://www.archive.org/details/colloidsinbiologOObech COLLOIDS IN BIOLOGY AND MEDICINE BY PROF. H. BECHHOLD Memher of the Royal Institute for Experimental Therapeutics in FranJ:fort A. M. AUTHORIZED TRAXSLATIOX FROM THE SECOND GERMAN EDITION, AVITH NOTES AND EMENDATIONS BY JESSE G. M. BULLOSA, A.B.. M.D. Assistant Clinical Professor of Medicine, Fordham University, Adjunct Professor of Clinical Medicine, Xeic York Polyclinic School and Hospital. Visiting Physician Riverside Sanatorium, Associate Visiting Physician, Willard Parker Hospital, New York City. 54 ILLZ'STBATIOXS NETV YORK D. VAX NOSTRAXD COMPANY 2ö Park Place 1919 Copyright, 1919, BY D. VAN NOSTRAND COMPANY B3g StanlJopc iprcss F. H.GILSON COMPANY BOSTON, U.S.A. AUTHOR'S . PREFACE. This book is an attempt to apply the results of colloid research to biology. The reader may find the undertaking somewhat bold, since the number of facts are so few, and the gaps so numerous that a complete picture is impossible. I find myself somewhat in the position of a palaeontologist who wishes to reconstruct the ancestry of the entire organized world from some chance fragments. Each day brings new finds which must be fitted into his plan and which confirm his views or show that he has been on a false trail. From the nature of things, it happens that I must more often indicate problems than report experimental results. This probably will prove a stimulus to those who wish to take active part in the de- velopment of our young science. I wish to state one additional fact: it was not my purpose to make this book exhaustive. I have endeavored to give a general view, and since the work is addressed to biologists and physicians as well as colloid investigators, I have striven to give a clear picture of the subject, disregarding moot questions. — On this account the "Introduction to The Study of Colloids" was abbreviated as much as possible without danger of obscuring the subsequent parts. — Those who wish to stud}^ more thoroughly pure colloid chemistry, I refer to the excellent books of Herbert Freundlich "Kapillarchemie" Leipzig, 1909, and Wolfgang Oswald "Grundris der Kolloidchemie," Dresden. Accordingly, in the arrangement of the first part, I have not fol- lowed the usual system, but have been guided by a desire to make easy of comprehension the matters most important to biologists and physicians. On this account I have considered it advisable to de- vote considerable space to the "Methods of Colloid Research." Some new unpublished experimental data of my own and some placed at my disposal by others have been included. It is finally my privilege to thank all those who have helped me in the preparation of this book, particularly Professors H. Apolant, R. Höber, H. Sachs, and Dr. H. Siedentopf. I am especially in- debted to my dear friend, Professor Richard Lorenz, with whom I have discussed some of the chapters, and to Dr. and Mrs. Ziegler, iv PREFACE who undertook not only the laborious task of reading the proofs but also checked back the index, page for page. I am also grateful to my publisher, Mr. Theodor Steinkopff, who, by intelligent coopera- tion was of great assistance to me. H. BECHHOLD. Frankfort on the Main DEDICATED TO HIS EXCELLENCY PRIVY MEDICAL COUNCILLOR PROF. DK PAUL EHRLICH AND PRIVY SANITARY COUNCILLOR DR. THEODORE NEÜBÜRGER WITH THE GRATITUDE OF THE AUTHOR TRANSLATOR'S PREFACE The translation of the first edition of this book was in hand when Professor Bechhold announced the preparation of a second edition. The translation was therefore delayed awaiting that issue in order to bring the volume up to date, and embody data obtainable from recent literature. The proof sheets were received in 1915 and 1916, but the transla- tion and revision were again delayed by occupation connected with the prosecution of the War. (Local Board Membership.) The translator hopes that Professor Bechhold's presentation of the colloid chemical problems of biology and medicine will serve to stim- ulate greater interest in colloid chemistry among physicians, biolo- gists, and bio-chemists. His own reward, he has already found in the interest attached to interpreting the every-day problems of prac- tice from this angle. It is to be expected that the application of colloid principles to the phenomena of life will lead to new and more rational theories of their disturbances, and it is hoped that an improved practice of medicine will eventually result The pages of this book indicate the begin- nings already made and suggest future studies. The translator acknowledges his debt to Dr. B. Michaelovsky for assistance with the translation; to Mrs. D. D. Pool for the author index, and to his publishers, Messrs. D. Van Nostrand Company, for their friendly co-operation. His debts to his friend, Jerome Alexander, he is unable to enu- merate. He owes to him, not only his introduction to colloid chem- istry, but never failing stimulation, encouragement and assistance with the translations and the proof-reading: Any merit the work possesses is due to his collaboration. J. G. M. B. February 15, 1919. 62 West Eighty-Seventh Street, New York City. THIS TRANSLATION IS DEDICATED WITH AFFECTIONATE GRATITUDE TO MY SISTER EMILIE M. BULLOWA, L.L.B TABLE OF CONTENTS PART I INTRODUCTION TO THE STUDY OF COLLOIDS Page Introduction xv CHAPTER I What are Colloids? 3-9 Sols — Suspension, emulsion, solution — Gels — Structure of jellies. CHAPTER II Surfaces 13-33 Surfaces — ■ Chemical combination — Solution — Adsorption — Sur- face pellicles. CHAPTER III Size of Particles, Molecular Weight, Osmotic Pressure, Conduc- tivity 41 CHAPTER IV Phenomena of Motion , 49-56 Brownian Zsigmondy movement — Diffusion — Diffusion in jellies — Membranes. CHAPTER V Consistency of Colloids 64-73 Internal friction — Swelling and shrinking — Life curve of colloids. CHAPTER VI Optical and Electrical Properties of Colloids 75-87 Optical properties — Electrical properties — Salting out — Floccula- tion — • Radioactive substances as colloids. CHAPTER VII Methods of Colloid Research 89-126 Dialysis — Ultrafiltration — Apparatus — Pressure — Gauging of ultrafilters — Adsorption by filters — Applications — Diffusion in aqueous solution — -Diffusion in a jelly — Osmotic pressure — Os- motic compensation method — Surface tension — Adsorption — In- ternal friction — Melting, coagulation and solidification temperatures xi Xll TABLE OF CONTENTS Page — Swelling — Flocculation — Electric migration — Optical methods — Interferometer — Ultramicroscopy for colloidal solutions — Ultra- microscopy for organized material. PART II THE BIOCOLLOIDS Introduction 129 CHAPTER VIII Carbohydrates 133 CHAPTER IX Lipoids 139 CHAPTER X Proteins 142-166 Albumins — Electrolyte-free albumin — Acid albumin — Alkali albumin — Albumin and inorganic hydrosols — - Albumin, heavy metals and salts of heavy metals — Globulin — Fibrin — Nucleins — Albuminoids — ■ Nucleo albumins — Hemaglobin — Colloid cleavage products of proteins. CHAPTER XI Food and Condiments 168-179 Meat — Milk and dairy products — Honey — Flour, dough and baking products — Beer. CHAPTER XII Enzymes 182 CHAPTER XIII Immunity Reactions 193-211 Nature of antigens and immune bodies — Distribution of immune substances between suspensions and solvent — Specific adsorption — Adsorption by organized suspensions — Distribution of immune sub- stances between dissolved colloids and solvents — Precipitation of dissolved colloids and organized suspensions — Electric charge, H and OH ions — ■ Complement fixation and Wassermann reaction — Wassermann reaction — Anaphylaxis — Protective ferment — Meio- stagmin reaction. PART III THE ORGANISM AS A COLLOID SYSTEM Significance of the Colloidal Condition for the Organism 213 CHAPTER XIV Metabolism and the Distribution of Material 215-245 Distribution of water in the normal organism — • Pathology of water distribution — Edema — Inflammation — Salt distribution — Cir- TABLE OF CONTENTS xiü . •■' . Page culation of material — Circulation of water — Circulation of water in animals — The movement of water in plants — Circulation of crystal- loids — Circulation of coUoids — Influence of membranes upon the interchange of substances — Assimilation and dissimilation. CHAPTER XV Growth, Metamorphosis and Development 252-274 Growth — Genesis of structures — ^ Layered structures — Biological growth — Ossification processes — Diseases of bone — Concrements — Gout. CHAPTER XVI The Cell 276-279 Protoplasm — Nucleus — Cell membrane and plasma peUicle. CHAPTER XVII The Movements of Organisms 282-292 Movements of lower organisms — Movements of higher organisms — Muscle as a colloid system — Muscle function (including the heart). CHAPTER XVIII Blood, Respiration, Circulation and Its Disturbances 299-315 Blood — Plasma — Lymph — Blood corpuscles — Respiration (gas exchange) — Circulation and its disturbances — Secretion and ab- sorption. CHAPTER XIX Absorption 317-323 Alimentary absorption — Parenteral absorption. CHAPTER XX Secretion and Excretion 326-345 Glands — Sahva — Bronchial glands — Gastric juice — Secretions which pour into the intestines — Ividneys and the secretion of urine — Concentration of the glomerular filtrate — Pathology of urine secretion — Result of deficient kidney function upon the organism — Urine, normal — Urine, pathological — Sweat glands — Milk. CHAPTER XXI The Nerves 352-356 Nerve irritability and swelhng — Cerebrospinal fluid — Integument — Internal secretions. PART IV CHAPTER XXII Toxicology and Pharmacology 359-416 Cooperation of indifferent substances — Colloids — Adsorption ther- 3'Py — • Colloidal metals — Action on microorganisms — Ferments — Xiv TABLE OF CONTENTS Page Autolysis — Metabolism — Temperature curve — Distribution — Therapeutics — Animal experiments — Clinical experiments — Mer- cury — Sulphur — Phosphorus, arsenic, antimony — Salts — Iron salts end iron oxid hydrosol — Narcotics and anesthetics — Disin- fection — Microorganisms — Methods of testing disinfectants — Diuretics and purgatives — Purgatives — Astringents — Balneology — Water and solutions — Salves and hniments. CHAPTER XXIII Microscopical Technic 417-435 Maceration and isolation — Fixing and hardening — Staining — Theory of staining — Technic of staining — The tissue elements in their relation to fixatives and dyes — Staining of bacteria. Author's Index 437 Subject Index 457 PART I. INTRODUCTION TO THE STUDY OF COLLOIDS. COLLOIDS IN BIOLOGY AND MEDICINE CHAPTER I. WHAT ARE COLLOIDS? In spite of the fact that they have a much wider distribution than crystalloids, it is only a little over fifty years that colloids have been scientifically studied. Plants and animals and all the things we manu- facture from them, such as our clothing and the greater part of our household goods, are colloids. In the year 1861, Thomas Graham,* ^ an Englishman, called attention to the fact that there were substances which, when in solution, diffused through parchment membranes (dia- lysed). These he called crystalloids because the soluble crystallizable substances (e.g., sugar and salt) possess this property to a marked de- gree. Substances which were held back by parchment membranes he called colloids, "glue-like," because glue was the most character- istic of this group. Every discoverer of a new fundamental principle is easily led into exaggerations; it so happened with Graham who opposed crystalloids to colloids as 'Hwo distinct worlds of matter," though we know now that all sorts of transition stages exist. In succeeding years very few investigators concerned themselves with colloids. The very fruitful development of organic chemistry occupied the attention of investigators, who neglected, as less important, a field which promised fewer immediate results. Only in the beginning of the new century was there a revival of interest in colloid chemistry. We shall not follow the historic development further, but shall give a description of colloids in accordance with the present state of the science. It must be noted at the outset that, even to-day, the behavior of a dissolved substance towards a partitioning membrane, that is, inability to diffuse through it, is the chief characteristic of a dissolved colloid. Many colloids form with liquids, especially with water, a more or less fluid solution. [The term dispersion is preferred by Arthur W. Thomas in a recent discussion on Nomenclature. Science, N. S., Vol. XLVII, No. 1201, p. 10. Tr.] This solution is called a sol (from 1 An * after an author's name refers to the reference in the index of authors. 3 4 COLLOIDS IN BIOLOGY AND MEDICINE solutio = solution). We speak of silver sols, albumin sols, etc. The dissolved substance in a sol may by various means be separated in an amorphous form that retains more or less water. This form is called a gel ^ (from similarity to gelatin) . If we add salt to a solution of colloidal silver, we obtain a black sediment containing very little water, the silver gel. If we boil a serum solution, the entire mass solidifies to a jelly that does not allow a separation of water and albumin, the albumin gel. Sols. As the researches of Graham have already shown, sols in general are substances which subdivide in their solvents into relatively large particles, or which possess very large molecules, so large that, in contrast with the molecules of water or crystalloids, they are un- able to pass through the pores of an animal skin or a parchment membrane. Chemical grounds indicate that albumin possesses a very large molecule. Even though we were to assume that it split into single molecules in aqueous solution, these are so large that they are unable to pass through an animal or vegetable membrane. Accordingly, the intact membranes of the organism protect it from loss of albumin; only in pathological conditions as in diseases of the kidneys does albumin pass through. Substances like albumin,^ soluble starches, etc., are to a certain extent inherently colloids. Every further subdivision of the colloid- ally dissolved particles would have to be associated with a sphtting of the molecule, and the fragments are certainly no longer albumin, but albumoses, Polypeptids, amino-acids, etc. It is otherwise in the case of certain artificial colloids. Accord- ing to G. Bredig and Th. Svedberg, gold, silver, platinum and other metals may be electrically pulverized under water or in organic fluids {e.g., isobutyl alcohol). According to G. Wegelin, sihca, vanadic acid, and other substances may by mere trituration be reduced to suspensions whose particles are so small that they can- not be recognized microscopically. If the electrical pulverization is accomphshed in water which is practically free from electrolytes, we obtain a solution which is red in the case of gold, brown in the case of silver, and greenish black with platinum. These solutions remain 1 Many recent authors make "gel" and "jelly" synonymous. It seems preferable to me to use the expression "gel" for the general comprehensive phenomenon and to reserve the word "jelly" for the gelatinization of a hydro- phile colloid. " When I refer to albumin, I mean albumin absolutely unsplit, whether it be egg albumin or globulin, etc., as opposed to albumoses which are classified as albumins in some textbooks. WßAT ARE COLLOIDS 5 unchanged for months provided they are preserved in Jena glass, which yields no electrolytes to water. These colloidal gold, silver or platinum solutions consist of more or less fine metal particles, each of which often comprises thousands of metal molecules. By varying the strength and tension of the current, finer or coarser particles may be obtained. Gold, silver and other sols have been prepared from gold, silver and other salts by chemically liberating the metal. According to the method of preparation, the metal is obtained in a more or less fine state of subdivision. If the metal sol is once produced it is impossible by the solvent alone to make the particles still smaller (without using chemical means). Unlike albumin they do not have the tendency to disintegrate of themselves in the solvent. We can accordingly call them artificial colloids, because they can be brought into such fine subdivision only by artificial means. If one were to further subdivide the molecules of such artificial colloids, the mole- cule would remain intact; gold would remain gold, and silver, silver. R. ZsiGMONDY * has prepared gold solutions so finely subdivided that they approach molecular dimensions, and Th. Svedberg has shown that, with these gold sols, and also with selenium sols, the finer the subdivision, the nearer these substances approach in color and light absorption their respective molecular solutions. In general, however, the artificial sols which can be seen in the ultramicroscope consist of much coarser particles than the natural sols. While the chemical constitution of inorganic colloids is revealed by their method of preparation, nothing was known concerning the constitution of natural organic colloids until 1913, when Emil Fischer succeeded in synthetically preparing organic colloids resembling tannin and having molecular weights above 4000. Exact knowledge of the chemical constitution of these substances wiU reveal much to colloid research. Suspension, Emulsion, Solution. By suspension we mean the floating of a powder in a fluid, e.g., clay in water. An emulsion is the minute division of one fluid in another with which it does not mix, e.g., oil in milk or water. The smaller the particles of the "dispersed phase" ^ (cf. p. 11) of the clay or the fat, the longer it takes for them to separate. Such a suspen- sion or emulsion, in which the dispersed phase is easily distinguished ^ Portions of a structure separated from each other by physical surfaces are called phases (Wilh. Ostwald). A mixture of oil and water contains two phases. Oil is one phase and water the other. Dispersed means scattered, distributed. In the above examples oil or clay is the "dispersed phase." 6 COLLOIDS IN BIOLOGY AND MEDICINE microscopically, may last months and even years. Only about a decade ago there was an animated discussion as to whether the known inorganic colloids such as colloidal silver, gold, arsenic sulphid, Prussian blue, etc., were suspensions or true homogeneous solutions. Some evidence was against their being considered homogeneous solutions; other evidence, however, favored this view. Microscopi- cally, they seemed entirely homogeneous, and they could not be separated from their solvents by mechanical means (filtration or centrif ugation) . It was only by means of the ultramicroscope, invented by H. Siedentopf and R. Zsigmondy (1903), that it was convincingly shown that they were suspensions and not homogeneous solutions. After this point was settled, the further question arose as to whether gelatin, albumin sol and like substances were to be con- sidered true solutions. Under the ultramicroscope, they also showed tiny particles, which, however, were by no means as numerous as might have been expected. Evidently most of them were invisible, and it was uncertain whether this was due to conditions of refrac- tion, or whether the larger part of these substances was in true solu- tion. This question was settled by the method of ultrafiltration invented by H. Bechhold in 1906. By a sufficiently impermeable jelly filter {ultr a filter) , that is, by a purely mechanical process, he was able to separate solutions of albumin, gelatin, enzymes, toxins, etc., from their aqueous solvent. Not only did albumin, gelatin, etc., prove to be suspensions or emulsions, but in addition, substances whose true solubility had hardly been questioned, e.g., most of the albumoseS; and even dextrin whose molecular weight had been placed at about 1000, and which had practically been classified as a crys- talloid. It is also possible to accomplish such a separation by means of centrif ugation. By centrifugation at 6,000 revolutions per minute, H. Bechhold separated colloidal silver sols (collargol) into coarser and finer particles. H. Friedenthal has recently constructed centrifuges turning from 10,000 to 30,000 revolutions per minute, by means of which he can separate the casein from cows' milk. We must here refer to the definition for "homogeneous" and "homogeneous solution" given by H. W. Bakhuis Roozebom, whose premature death we lament: "We call a system homogeneous, if all its mechanically separable particles possess the sarne composition and the same physical properties. Therefore, this homogeneity of con- stitution exists in a well-mixed liquid only because of the smallness of molecules and the coarseness of our means of observation." We cannot speak of a definite solubility in respect to suspensions WHAT ARE COLLOIDS 7 and emulsions, for within certain limits, we are able to suspend as much clay or emulsify as much fat as we wish; the ''finer" the clay or the fat is subdivided, the more "dissolves." The same thing holds for colloids, which are characteristically different in this re- spect from crystalloids, the latter having a sharply defined solubility. As a matter of fact we can get "supersaturated solutions" of crystalloids, and certain small additions increase the solubility dis- proportionately. Such additions (e.g., albumin, albumoses, gelatose, dextrin) when employed in the case of suspensions and colloids, are called protective colloids (schutz-kolloide) because they protect the lixiviated clay or finely dispersed silver from separating out. As indicated, many of the pure, inorganic sols, especially the metal sols obtained by electric pulverization, are very sensitive to elec- trolytes by which they are easily precipitated, whereas on the con- trary natural colloids are relatively insensitive. It has been shown that the addition of certain natural sols acting as protective colloids gives metal sols, etc., properties which cause them to approach the natural sols in stability. The inorganic colloids employed in medi- cine, such as colloidal silver (collargol, lysargin), colloidal calomel (kalomelol), colloidal bismuth, etc., are all stabilized by protective colloids. Thus we see a complete transition from the suspension and emul- sion of insoluble substances, to the true solution of crystalloids, where there occurs a disintegration by the solvent, which is so profound in the case of electrolytes, that they separate into their electrically charged atoms (ions). As everywhere in nature, here too there are no sharp lines of demarcation. We cannot deny that at a certain size the particles possess the maximum colloidal properties, especially those conditioned by surface phenomena. These properties decrease when the particles are larger, i.e., if they approach those of true sus- pensions or emulsions; or when they become smaller, i.e., if they approach the molecular condition. Th. Svedberg* has shown that the light absorption of colloidal gold and selenium increases as the particles become smaller, reaches a maximum in the amicroscopic field, and again decreases as the particles approach molecular dimensions. It is noteworthy also that at a certain degree of dispersion the tinctorial power reaches a maximum which in the case of gold is forty times stronger than the powerful color fuchsin. The color of colloidal gold having a particle size of 10 to 20 nn is ruby red; when the particles are smaller it is fuchsin red; but when the particles are still smaller the color becomes yellowish red. In other words it approaches the color of gold salts (auric chlorid) in which the gold is molecularly dispersed. 8 COLLOIDS IN BIOLOGY AND MEDICINE To recapitulate: The chief characteristic of sols is the large size of their particles/ which are unable to pass through vegetable or animal membranes. The natural size of the particles accounts for this in the case of natural sols, while in artificial sols it is due to the defects of our technic, which hitherto has not permitted our prep- aration of such substances in molecular, or even approximately- molecular subdivision. This criterion is only valid for extreme cases. Between the un- doubted colloids, e.g., albumin, and the undoubted crystalloids, e.g., amino acids, there are all kinds of transition forms, which pass through the same membranes more or less rapidly, e.g., albumosea and peptones. There is, indeed, no sharp line of demarcation be- tween colloids and crystalloids. Gels. It might be inferred from the nomenclature (colloids and crys- talloids) that the main distinguishing feature was the ability or inability to crystallize. It is a fact that most crystalloids, i.e., sub- stances which pass through membranes, are cry stalliz able, whereas most colloids are not able to form crystals when they separate from solution. However, this is not a radical difference, since egg albumin and hemoglobin which are undoubted colloids may be obtained in beautiful crystals; and I have further estabUshed by ultrafiltration, the colloidal nature of the solutions of the alkaline salts of the fatty acids (e.g., oleic acid) which also form good crystals. Colloids usu- ally separate from their solution in unformed masses called gels. If the solid phase be separated from crystalloid solutions, it may- form either crystals or a slightly or even an entirely amorphous pre- cipitate. Cuboidal crystals separate from a solution of common salt on evaporation or addition of alcohol, and crystals of Na2S04 + 10 II2O separate from solutions of Glauber's salt (sodium sul- phate). A white precipitate of barium sulphate, which has hardly any definite form, separates from a sodium sulphate solution upon adding barium chlorid. A substance of constant chemical com- position especially as regards water content is obtained if the im- purities, especially the extraneous water, are removed by filtration from the crystals or precipitate. To return to our example: the sodium chlorid crystals and the barium sulphate are water-free, whereas the sodium sulphate contains 10 molecules of water to one molecule of Na2S04, but it is water-free above 33° C. 1 By the passage through a membrane, I mean especially passage by means of dialysis. In many cases we may substitute ultrafiltration, provided ultrafiltera of sufficient tightness are employed (see p. 102). WHAT ARE COLLOIDS 9 Gels behave differently; in fact there are a number of colloids which separate from their solution almost water-free. If sols of gold, silver, platinum, arsenious sulphid or antimony sulphid hydro- sol, prepared according to the method of Bredig or the method of SvEDBERG, precipitate, that is, separate from their solutions in the form of flocks, they are almost free from water. Many inorganic sols {i.e., the artificial ones), and according to my knowledge nearly all natural organic sols, retain a large quantity of water upon separation. Gelatin is the most characteristic gel; its aqueous solutions (con- taining only 1 per cent of water-free gelatin) gelatinize at ice-box temperature. Furthermore, other sols, such as egg albumin, starches, silicic acid, iron oxid, etc., on separation in gel form retain many times their own weight of water, and form jelly-like masses in which the proportion between the solid subsiance ana the retained solvent is by no means constant. According to the circumstances attending the separation, the amount of water held fast in the gel has wide limits of variation. This is a cardinal distinction. In accord with it, I have adopted the happily chosen nomenclature of J. Perrin and call such colloids as throw down a practically water-free hydro- gel, hydrophobe and those which produce a hydrogel swollen and rich in water, hydrophile colloids.^ The gels of hydrosols (cf. p. 11) stabilized by protective colloids are somewhat hydrophile, because very minute quantities of pro- tective colloid are sufficient to give the inorganic sol the properties of the protective colloid. The Structure of Jellies. Jelhes are formed from their respective solutions by such physical and chemical changes as would cause the separation of crystals in the solution of a crystalloid, e.g., by cooling, by removal of water either by a chemical change or by forming an insoluble substance {e.g., by boiling or by acidifying an albumin sol) . It is thus apparent that jellies are to be considered two-phased structures. Two phases are much more obvious in coagulated egg albumen whose opacity and white color suggest a non-homogeneity of struc- ture. Recently Bachmann * has demonstrated ultramicroscopically the two-phased structure of transparent jellies such as gelatin and silicic acid. In gelatinizing, it is evident that granular flocculent 1 I mentioned above that as far as I knew all natural organic sols are hydro- phile. It might be objected that epidermis, hair, feathers, bark and numerous other vegetable structures are deposited from natural sols and become very poor in water. This is met by the assertion that when they are deposited they contain much water and that the loss of water or drying out occurs subsequently. 10 COLLOIDS IN BIOLOGY AND MEDICINE and even crystalline particles, e.g., in soap jellies, unite to form spongelike structures. J. M. Van Bemmelen compares the process of separation of colloids with the condition , deduced from the phase rule govern- ing the separation of two fluids not miscible in all proportions, e.g., water and phenol. In jellies also, we have a phase containing much colloid and little water, and another phase containing little colloid and much water. This conception of the structures of jellies and the process of separation is due to J. M. Van Bemmelen, O. Bütschli, W. B. Hardy, G. Quincke, R. Zsigmondy and W. Bachmann. The views of O. Bütschli are not accepted nowadays. He main- tained that jelHes are, broadly speaking, foamy structures having microscopic cavities with firm net-like walls filled with fluid. Such a structure can occur only exceptionally. The conception of jelhes as spongy structures gives us a satis- factory explanation of their properties. It explains their solidity and their plasticity, their elasticity, and in short their various physi- cal properties. The above assumption finds corroboration in another observation, through the fact that jellies also act as ultrafilters and consequently must be penetrated by fine capillaries whose diameter has been determined by Bechhold (see p. 99). The penetration of jellies by fine capillaries filled with fluid was demonstrated by another observation. H. Bechhold and J. Ziegler* allowed salts which would form precipitates with various properties to diffuse towards each other in gelatin, e.g., potassium ferrocyanid and copper sulphate which form a copper ferrocyanid membrane entirely impervious to electrolytes; silver nitrate and sodium chlorid which form a silver chlorid membrane which is permeable for electrolytes if the osmotic pressure is higher on one side than on the other. Microscopic sections through the membranes formed by the precipitates prove that the gelatin is not deformed. Accordingly, when diffusion ceases, it is because the diffusion paths have been obstructed, i.e., a precipitate has been formed in the fluid phase so that the paths are closed and of course the gelatin walls which contain little water are impassable for electrolytes. Remelting is sufficient to reopen the diffusion paths. Anderson determined a diameter of 5.2 fxfj. for the largest pores of silicic acid jelly from the vapor pressure reduction which a fluid undergoes in cylindrical capillaries. Until now we have assumed that sols exist only a§ aqueous solu- tions and that there are only aqueous gels. This is by no means true. Even Thomas Graham * showed that water could be replaced by alcohol and glycerin. Th. Svedberg pulverized numerous metals WHAT ARE COLLOIDS 11 in organic fluids especially in isobutyl alcohol. R. Lorenz has even made metal sols in red hot solution (pyrosols) by electrolysis of molten lead and cadmium salts, etc. To distinguish them from the water soluble hydrosols and from the hydrogels we call them organo- sols or organogels and according to the solvent as alcosols, etc. These do not occur in nature and are therefore of no importance to us. An investigation of such colloids as have fats, lecithin and Cholesterin either as a dispersing medium or as a "dispersed phase" (cf. p. 12) would certainly be of great importance to biology and medicine. The fact that fats and oils, especially mineral oils, may serve as a dispersing medium for colloids has been mentioned by D. Holde * in his work on the "physical condition of solid fats" and this was confirmed by investigations of H. Bechhold,i who was able upon ultrafiltration (through a toluol-glacial acetic acid collodion filter) to hold back from a crude oil a part of the asphalt colloidally dis- solved in it. Recently C. Amberger dissolved a series of metals (gold, silver, platinum, arsenic, etc.) in lanolin. Some of these solutions have therapeutic application. By ultrafiltration H. Bechhold separated from commercial chlorophyl the coloring matter and the wax-like products which were evidently held in colloidal solution. Protective colloids also exist for organic fluids. Iron oxid gel and iron oxid hydrosol, rennin and trypsin, as well as albumoses which are completely insoluble in chloroform, become soluble in it with the aid of lecithin acting as a protective colloid. Thus far we have sought to obtain a picture of what we term ^'colloids" and now we shall strive to elucidate upon what their properties depend. In solutions of cofloids and in gels, we have mixtures of solids or of fluids with fluids. It has for a long time been known, that at the interface between two substances which do not mix (air and water, oil and water, glass and water) there occur phenomena, called surface phenomena. For instance, the surface of water in contact with air acts as a pellicle; if we allow water to drip, each drop reaches a considerable size before its weight breaks through the surface skin or pellicle and the drop falls. This surface skin is much weaker in the case of alcohol, so that drops of alcohol falling from the same tube are much smaller than those of water. The fol- lowing is another example of a surface phenomenon: oil forms a sphere in a suitable mixture of water and alcohol; if we raise the specific gravity of the water by removing some of the alcohol, the oil rises and spreads out over the surface of the water. Such surface phenomena are very numerous; they are brought 1 Unpublished. 12 COLLOIDS IN BIOLOGY AND MEDICINE about by the fact that different conditions exist in the interior than exist on the surface of the fluid or sohd body. In two-phase systems, such as colloids, in which the interfaces reach enormous dimensions, surface and capillary phenomena become most promi- nent; in fact, they are in many ways the most characteristic phe- nomena of colloids. Two fluids which do not mix are described as two fluid ''phases"; it is possible also to speak of a fluid, of a solid and of a gaseous phase. In order to give expression to the great surface development in a sol or gel, Wolfgang Ostwald introduced the very happy expression dispersed phase. In a silver sol, silver is the soUd dispersed phase; in an oil emulsion, oil is the fluid dis- persed phase; in both, water is the dispersing medium. Colloidal solutions and gels are all dispersed systems. [Of interest in this connection is the work of G. H. A. Clowes and Martin H. Fischer. Tr.] CHAPTER II. SURFACES. We have seen in the preceding chapter that both colloidal solutions and jellies are to be regarded as two-phase systems. In dispersed systems, the surfaces of contact acquire an overpowering importance by reason of their enormous development. In order to get an idea of the increased development of surface attending progressive subdivision, I give below a table taken from Wolfgang Ostwald: Edge length Number of cubes occupying a vol- ume of 1 cm.3 Total surface 1 cm 0.1 cm 0.01 cm 1 10^ 10« 10« 1012 1015 1018 1021 1024 6 cm. 2 60 cm. 2 600 cm. 2 0.001 cm. (The diameter of a human blood corpuscle is about 0.0007 cm.) 6,000 cm. 2 Im 0.1m ( = 0.0001 cm.; diameter of a small coccus). 6m.2 60m.2 0.01m 1mm (Limit of ultramicroscopic visibility) 600 m.2 (The diameter of the finest colloidal parti- 6,000 m.2 60,000 m.2 0.1mm (Diameter of elemental molecules) From this, we can understand how small surface forces may, in a dispersed system, become most important, and mask other phe- nomena. We shall proceed to the study of the properties of sur- faces in the following pages. If we compare a point A (Fig. 1) in the interior of a phase, e.g., a fluid, with one on the surface A', e.g., in contact with air, we notice that the former is surrounded on all sides by an impervious mass, whereas the latter, surrounded on only one side by such a mass, experiences an attraction to the fluid phase. The pressure with which the surface layer is drawn inwards is called inward attraction (Jbinnendruck) . If we imagine a drop of water to be on a surface which it does not moisten, e.g., a leaf, there is a pulling towards the center from all sides, which means that the drop takes a spherical form, in order that it may have the smallest possible surface. The surface acts 13 14 COLLOIDS IN BIOLOGY AND MEDICINE like an elastic skin which surrounds the drop. If we drop water from a tube, the drop develops in size and is retained by its skin until the increasing weight tears it away.^ We call this force which determines the tension of the surface, the surface tension (a). A flat surface of water of 1 square cm. endeavors to contract with a tr of about 0.075 gm., when it is spread out like a soap bubble with air Air interface Fluid Fig. 1. pressure. Every change in shape of the sphere of water, i.e., every increase of the surface, presupposes work; this depends upon the surface area (co) and the surface tension (a). Surface energy = a ' (ji surface energy dynes or (7 = ^^ expressed m CO cm. There are many methods for determining the surface tension. Some of them depend upon the shape (distortion) which the sur- face of a fluid takes; some, upon the height to which a fluid ascends a capillary; some, upon the determination of the maximum weight attained by a falling drop. Detailed descriptions are to be found in every large treatise on Physics, such as H. Freundlich's " Kapillarchemie." A number of values of a are given as examples. The a of water is determined most frequently; it has the highest a of all the sub- stances which are fluid at room temperature (mercury excepted). The various methods give quite divergent values. The values given here are in each case the surface tension towards air. (fluid/air) (T water/oil + a oil/air 75 > 22.9 + 32.7. 1 Bibliography in Wo. Ostwald, " Handbook of Colloid Chemistry," trans, by Fischer, Oesper and Berman, Phila., 1915. 18 COLLOIDS IN BIOLOGY AND MEDICINE Expressed in general terms, this means that a fluid 2 spreads itself on the free surface of a fluid 1, if O"! >0'2 + 0'l/2- (Ti = surface tension of fluid 1 acting against air. 0-2 = surface tension of fluid 2 acting against air. (Ti/2 = surface tension of fluid 1 acting against fluid 2. Similarly, a fluid 3 spreads over the common boundaries of two fluids 1 and 2, whenever 0'l/2 > ö'2/3 — (Ts/l. This phenomenon is of the greatest biological interest, because it follows that many fluids must spread out on the boundaries of other fluids or solid bodies and form films, and further, that solid particles, suspensions and colloids must collect on surfaces, according to con- ditions. The diminution of surface tension between two surfaces is the forerunner of mixing or solution; there exists no surface tension between two readily miscible fluids. We shall discuss the distribu- tion of dissolved substances on interfaces more fully on page 33 in the discussion of surface pellicles. We shall see that numerous organized structures and indeed the movement of protoplasm and of the lower animals are derived from this phenomenon of surface tension. It is the monumental service of G. Quincke that he showed the connection between the purely physical processes and the phe- nomena of animate nature. The surface tension of solids is deduced from the fact that finely divided particles are more easily and rapidly dissolved than coarser ones; it also explains the fact that artificial hydrophobe colloids are produced only in a dispersing medium in which they are wholly in- soluble (see p. 73). The slightest solubility permits them to pass from the dispersed phase into coarser particles. Only to a limited extent is it possible directly to measure the surface tension of solid bodies. Roentgen measured the a for rubber/air and rubber/water and Tangl * tested a new method on the interfaces of rubber/water and paraffin/ water. The basis for the method is that a tube of the substance to be tested (rubber or paraffin) undergoes a change in shape when it is plunged from the air into a fluid (water) . Interfaces of Solutions. Whenever substances are dissolved in one or both phases, there is usually a difference between the con- centration on the surface and on the interior. These changes in concentration at the surface are termed adsorption. A substance be- comes concentrated upon the surface if it reduces the surface tension. This is the most usual adsorption phenomenon. Only a few inor- ^ SURFACES 19 ganic salts increase the surface tension of water and they are, ac- cordingly, less concentrated at the surface than in the interior of the solution. This latter occurrence, negative adsorption, is of significance for the distribution of salts in cells, as will be shown on page 25. ''Adsorption," which is of the greatest significance in colloid research, will, in the subsequent paragraphs, be considered from the standpoint of the distribution of a dissolved substance between two phases. In this connection also reference should be made to page 33 (surface pellicles). CHEMICAL COMBINATION, SOLUTION, ADSORPTION. We have seen that colloids are diphasic systems, and the ques- tion must arise as to how a third substance will he distributed between the two phases.^ Chemical Combination. If we take as the dispersed phase, a suspension of calcium carbon- ate (precipitated chalk) in water, this will represent the solution of a hydrophobe colloid. If we add sulphuric acid to the suspension, the acid will be immediately and completely bound. It is impossible to detect free sulphuric acid in the suspension by any reagent if any calcium carbonate still remains in the suspension.^ If we continue with the addition of sulphuric acid, a point will suddenly occur when, no matter how much is added, sulphuric acid is no longer bound by the chalk; all the excess remains in the water. We are accustomed to say that a chemical reaction has occurred between the calcium carbonate and the sulphuric acid and that there is a chemical union of Ca and SO4. Ca unites firmly with a definite quantity of SO4. We may add as much water as we want; it cannot abstract any S04from the CaS04. The process is irreversible (cannot be reversed). Solution. If we emulsify carbon disulphid in water, and add a little bromin, the entire fluid is colored brown. If we allow the carbon disulphid to settle, the water is light brown and the carbon disulphid is colored ^ In this instance "distribution" is used in the most general sense, though in physical chemistry it is employed only to indicate the distribution of a sub- stance between two solvents. 2 Although the formation of salts on mixing dissolved acids and bases results with infinite rapidity, it takes an appreciable time in the case of colloid solutions (and of course with coarse suspensions). Vorländer and Häberle, 20 COLLOIDS IN BIOLOGY AND MEDICINE dark brown. The more bromin we add, the darker both the water and the carbon disulphid become; the latter, however, is always more darkly colored than the former. If we study the process quantitatively, the following becomes evident: if in a given case the concentration of the bromin in the carbon disulphid is c (carbon Til.,..,, ^ / j^ N ^1 cfcarbon disulphid) disulphid), m the water c (water), then -, — - — r-^ = n, that is, the relative distribution of the bromin in a given case is n. If we double the quantity of both carbon disulphid and of water and then test the quantity of bromin in the fluids, we shall find that in both the concentration has fallen to half and that the distribution continues to be n. If we double the quantity of water, its color is only slightly less intense, because bromin enters the water from the carbon disulphid. If we now measure the bromin content of the two fluids, we shall find ac (carbon disulphid) and in the water &c (water); that is, ac (carbon disulphid) _ 6c (water) No matter how we vary the quantity of solvent or of bromin, the apportionment of bromin is always n. We may, therefore, say that c n is a constant and express it — = k. Ci This equation is characteristic for the distribution of a substance between two phases in which it is soluble. The process is reversible; it is in labile balance. The law of distribution was formulated by M. Berthelot and Jungfleisch in 1872, though we still frequently refer to Henry's Law of Distribution. Strictly speaking, this ex- pression applies only to the distribution of a gas between a fluid and a gaseous phase (proportionately to the pressure). c . . The distribution — = k applies only in case the molecular weight of the dissolved substance is the same in both solvents. If this is c not the case, the equation becomes — = k, in which a and b express the difference in the molecular weight in the two solvents.^ W. Nernst has formulated the law of distribution in this way. For example, benzoic acid in water has a simple molecular weight, but in benzol it exists mostly in double molecules. In order to ex- press this, the equation of distribution becomes . =- = k. VC (benzol) 1 [Prof. J. L. R. Morgan suggests that a better form would be the following: — = k, where x is the ratio of the molecular weight in solvent (2) to that in solvent (1). Sx ^ xs and — -. — TTT- = X. Tr.j (2) (1) m m (1) SURFACES 21 Adsorption. With colloids in particular there occurs a third possibiUty of dis- tribution, wherein the surface comes into play rather than the total mass of the dispersed phase. The condition of distribution which we are about to describe is called adsorption. Suppose we suspend in water a substance which we may assume does not dissolve or undergo chemical combination, e.g., pure carbon. We know that bone black may to a greater or less extent decolorize dye solutions; that it is used to bleach dark sugar juices and to decolorize the dark solutions of the organic chemist. When suspension of powdered char- coal is added to bromin water, we observe the following: If we add very little bromin to the water the latter will become completely decolorized; if we add more, a considerable part is taken up by the charcoal but the water becomes brownish. With further addition of bromin the water is colored more intensely and the charcoal takes up proportionately less bromin. This process is reversible and the distribution of bromin between charcoal and water follows a cer- tain law. We cannot as in the case of a solvent, however, speak of the concentration of the dispersed phase. In experiments, it has become customary to insert the specific gravity of the dispersed phase, and this custom is, as a rule, justified. Let us, for example, designate by x the amount in millimols of bromin that is adsorbed from a solution by m grams of charcoal, and by c, the concentration of the bromin in the water after adsorption. If we deal with sub- stances of unknown molecular weight, x indicates the weight in milli- grams and c the weight which is present in 1 cc. of water after determining the adsorption balance. Empirically we arrive at the equation — (adsorbed) m T = /«, c" (free) in which the exponent - is always < 1. Inspection shows that if X n = 3 and k = 20, the equation is satisfied when — in the char- m coal = 200 and c (water) = 1000. If we dissolve but very little bromin, then the equation is satisfied X when, e.g.,— = 20 and c = 1. In the first instance only 1/5 of the bromin is adsorbed by the charcoal, but in very great dilutions 20 times as much goes to the adsorbent. 22 COLLOIDS IN BIOLOGY AND MEDICINE If it is unnecessary to determine constants, the direct graphic representation of results is the simplest method. The concentra- tion in water (the dispersing substance) is made the abscissa. The ordinate is the adsorbed amount of the dispersed phase. (This is the difference between the entire quantity of the substance dis- solved and what remains in solution.) The points of intersection are points of the curve experimentally derived. The lines show us at a glance (as is seen in Fig. 2), in simple cases, whether the dis- cs a, ^' ^' - / f / / ,^ / / / / / / 1 i i / / / / / j / / 1 / I / C (concentration) in the water Fig. 2. tribution and curve is one of chemical combination, solution or adsorption. The heavy continuous line ( ) is the graphic representation of a chemical reaction: 3 mols CaCOs are suspended in water and H2SO4 is added. It is at once seen from the diagram that 3 mols II2SO4 are bound, i.e., there is no free H2SO4 in the water, but on the addition of more II2SO4 the dispersed phase can take up no more acid, so that the acid remains in the dispersing medium. The broken line ( ) is the graphic representation of the dis- tribution of a substance between two solvents. The dot and dash line (- • -) is an adsorption curve. For the graphic representation of such adsorption phenomena the above equation is solved by log- arithms; and we obtain X 1 log — = log fc 4- - • log c. m n SURFACES 23 This is the equation of a Hne. If the logarithms of the values found for — and for c, of different concentrations of the substances under , ^ m examination, are carried onto a rectangular system of co-ordinates, these points would lie on a line, provided the substance was sub- jected only to pure mechanical adsorption. As the simplest example we shall choose the curves and data which H. Freundlich * derived from his studies of the adsorption of certain fatty acids by charcoal. Fig. 3 shows the adsorption curve Fig. 3. Adsorption of fatty acids by charcoal. (After H. Freundlich.) of acetic acid, propionic acid and succinic acid, in direct graphic representation as we have explained on page 22. They are derived from the following somewhat abridged data of Freundlich: Acetic acid. / millimolsx x / millimols \ / m \gni. charcoal/ 0.0181 0.0616 0.2677 0.8817 2.785 0.467 0.801 1.55 2.48 3.76 Propionic acid. / millimols \ x / millimols \ A cc. /TO Vgm. charcoal/ 0.0201 0.0516 0.2466 0.6707 1.580 0.785 1.22 2.11 2.94 3.78 Succinic acid. /miIlimols\ x I millimols \ V cc. / m \ gm. charcoal/ 0.0076 0.0263 0.0477 0.2831 1.161 1.09 1.70 1.95 3.26 4.37 24 COLLOIDS IN BIOLOGY AND MEDICINE Fig. 4 shows the hne passing through the logarithms of these data and it should be observed that all logarithms less than unity are negative. The tangents to the angle of inclination between the elements of Fig. 4. (After H. Freundlich.) the curve (acetic acid, propionic acid and succinic acid) and the abscissa (log c) is the exponent -. The distance on the ordinate log — ) from the zero point (origin of co-ordinates) to the point inter- secting the uniting lines is log k. They have these values: Acetic acid - = 0.425, k = 2.606. n Propionic acid - = 0.354, k = 3.463. Succinic acid - = 0.274, k = 4.426. n Since the values observed do not lie all in the same line, as is shown in Fig. 4, a mean value for the angle whose tangent is n may be de- rived by means of a protractor. In the same way, log k is not de- rived from the intersection of the last element of the curve, but from the mean value. The exponent - conditions the shape of the curve, and varies n within moderate limits. Though marked exceptions have been ob- served, it fluctuates usually between 0.5 and 0.8 as H. Freundlich has shown in his numerous experiments. The constant k in the adsorption formula is, in an ideal case, a natural constant which may be as characteristic for the adsorbed substance as k is in the distribution between two solvents. The great difficulty lies in the fact that, in the case of the dispersed adsorbing phase, we do not consider the mass, which may be easily SURFACES 25 ..<■ determined by weighing or measuring, but the surface which may vary greatly with the selfsame weight. That in fact not the mass but the surface of the dispersed phase is of importance in adsorp- tion, is evident from the following. Freundlich and Schucht* permitted dyes to be adsorbed by amorphous, i.e., colloidal, mercuric sulphid flocks. When the HgS became crystallin the dye dissolved again. We have here a counter- part of enzym action in which the adsorbed enzym (e.g., pepsin) is liberated when the adsorbing substrate (fibrin) changes its surface as it is split up. W. Mecklenburg* succeeded in obtaining different curves in the adsorption of phosphoric acid by colloidal stannic acid and of arsenic by ferric hydroxid when, starting with solutions of identical concentration, he precipitated stannic acid and ferric hydroxid at different temperatures; all the other conditions were identical. The lower the temperature at which stannic acid or ferric hydroxid gel were formed the more phosphoric acid or arsenic was adsorbed. The peculiarity of these curves was the similarity in their shape which is described by mathematicians as "affinitive"; on this account Mecklenburg called them affinitive adsorption curves. They can not be otherwise explained than that the same mass of adsorbent may have a different surface depending on the tempera- ture at which it was formed. Adsorption is a phenomenon which is conditioned hy the decrease of the surface tension of the solvent in respect to the dissolved sub- stance, at the interface between the solvent and adsorbent. In 1888, G. Quincke showed that substances which decrease the surface tension between the solvent and the dispersed phase must collect about the dispersed phase with the formation of a film. H. Freund- lich has elaborated and experimentally established the theory of adsorption phenomena, basing his ideas on Gibbs' Theorem.^ The marked diminution of the surface tension of water by fats, fatty acid, soaps, albumin and its cleavage products, and enzymes is characteristic; and it is not surprising that these substances are very easily adsorbed. From what has been already said, adsorption appears to be a purely physical phenomenon in which the chemical relations between adsorbent and adsorbed substance play no part whatever. [This 1 Gibbs' Theorem states: A dissolved substance is positively adsorbed if it depresses the surface tension, negatively adsorbed when it raises it. Willard Gibbs deduced these relations for gaseous mixtures and not for fluid solutions. The statement of W. Gibbs, that a small amount of a dissolved substance may powerfully depress surface tension but cannot raise it much, is likewise important (for further proof see H. Freundlich 's " KapiJlarchemie"). 26 COLLOIDS IN BIOLOGY AND MEDICINE statement is too dogmatic; in dealing with colloidal particles which approach molecular dimensions no sharp Hne can be drawn between physical and chemical forces. Tr.] We shall, therefore, with Wo. OsTWALD, call these purely physical phenomena mechanical adsorip- tion. Most of the investigations on adsorption have been conducted with pulverized solids, hydrophobe colloids, and gels as adsorbents. From a biological standpoint studies of adsorption by hydrophile sols are of especial importance, when we consider for example what occurs in the blood. The only investigations of this character that I am acquainted with are those of H. Bechhold*^ on the distribution of methylene blue between water and serum albumin. The volume in solution is directly obtained by ultrafiltration and the amount of methylene blue distributed in the aqueous filtrate thus obtained is measured. It was shown that in very dilute solutions of methylene blue the major part is held firmly by the albumin; whereas in greater concentrations the distribution is displaced in the direction of the water. The curve is similar to that of an adsorption curve. [The work of A. B. Macallum, " Surface Tension and Vital Phenomena," University of Toronto Studies. Physiol. Series No. 8 (1912) should be read in this connection as it involves the adsorption of potassium and its concentration at surfaces. Tr.] Based on what has been said hitherto, we might believe that nothing is easier than to determine by the curves of distribution whether we are dealing with chemical combination, distribution between two solvents or adsorption. If, however, we examine the experimental data, we see that there is rarely close agreement between observation and the calculated results. These divergences led to the derivation of other formulas deter- mined by the following considerations. According to the formula on page 21 the more concentrated a solution, the more is there adsorbed from it. Actually, saturation limits are reached in many cases. This may be explained as follows : Each interface can adsorb only a layer of a definite thickness, and saturation is reached when this layer is filled with adsorbed molecules. The formulas of Arrhenius,* Rob Marc,* and C. G. Schmidt* were made to be in conformity with the observed facts. The question, whether we are dealing with adsorption or solution, is frequently difficult to decide when the dissolved substance has a different molecular weight in the dispersed phase than in the solvent (see p. 20). The curve of distribution may then assume the form of an adsorption curve and in solving the problem all the incidental circumstances must be considered; for instance, W. Biltz*^ found SURFACES 27 ,.<■ that the distribution of arsenious acid between iron hydroxid gel and water follows the equation X — adsorbed 3- = 0.631. c (free)5 Were we to believe that the arsenious acid went into solid solution in iron hydroxid gel we should have to conclude that the arsenious acid had a molecular weight one-fifth as large in iron hydroxid gel as in water. From other observations, however, we know that arsenious acid in water breaks up into simple molecules so that the assumption that it dissolves in iron hydroxid gel is untenable. The following statements will show that many conditions may modify the course of the adsorption curve and such cases, according to L. Michaelis, are best called abnormal adsorption. A substance may, for instance, as the result of swelling, absorb more water than the substance dissolved and thereby simulate a nega- tive adsorption. Thus, B. Hekzog and Adler* found that talcum powder adsorbed from sugar or albumin solution more water than sugar or albumin so that as a result, the solution appeared to be more concentrated at the end than at the beginning of the experiment. The great conceiitration which adsorbed substances cause in the dispersed phase may be associated with changes in its condition; for instance, it may be thrown down as a solid. It has been ob- served, that charcoal which has been shaken with a solution of anilin dye shows the greenish metallic shimmer and the dichroism of the solid dye; albumin may coagulate at interfaces. Pro- found changes may accompany these variations of condition. This may be shown by the following examples, the so-called basic anilin dyes are salts consisting of a strong acid (usually hydrochloric acid) and a weak color base. The aqueous solution undergoes strong hydrolytic dissociation; the free color base shows a more or less colloidal character and is always strongly adsorbed. H. Freund- lich and G. Losev * showed that the color bases adsorbed by char- coal from new f uchsin and crystal violet were changed at the surface of the charcoal, and substances with entirely different properties were formed; probably they were the insoluble substances which A. VON Baeyer had previously prepared separately. In the case of the basic dyes, these chemically changed substances form the color on the textile fiber. Now we must recall that pure adsorption is a condition of equilib- rium changeable with the concentration of the dissolved substance. There can be no balance if a substance is removed from the solution 28 COLLOIDS IN BIOLOGY AND MEDICINE and thus made insoluble (irreversible) as in the above-mentioned ex- ample of the basic dyes. As dye stuff becomes insoluble, the char- coal or fibers must adsorb more dye and the process will continue until all the dye is adsorbed from the solution. In the case in ques- tion, the process is prematurely ended because of the concentration of the HCl hydrolytically separated, which to a certain degree ex- ercises a solvent action on the colored condensation product. Wilh. OsTWALD * has called attention to the fact that in solutions which are hydrolytically split into fatty acids and free alkali, there is a marked displacement of the adsorption balance. This occurs in washing. The fatty acids are adsorbed by the fabric and the skin; to accomplish this, there must occur in the solution a further hydroly- sis, i.e., a splitting off of alkali. In other cases the dissolved sub- stances may be completely removed from solution, e.g., albumin from urine (by charcoal, silicic acid or mastic emulsion, as adsorb- ents). If we try to determine the adsorption curves for the phenomena just described (especially the fixation of dyes), we shall find that the curve might be mistaken for that of an irreversible chemical process. Still more peculiar are the curves which W. Biltz and H. Steiner * obtained for the adsorption of night blue and Victoria blue by cotton, and H. Freundlich *^ for the adsorption of strychnine salts by charcoal or arsenic trisulphid, as well as G. Dreyer and J. Sholto * for the adsorption of agglutinins by bacteria. In these cases less substance was taken up by the adsorbent from the con- centrated solutions than from those of medium concentration. The explanation of a phenomenon of this kind was given by A. Lottermoser,* who observed with A. Rothe that amorphous silver iodid adsorbed less potassium iodid from highly concentrated solu- tions than from solutions of medium concentration. The process is influenced by the fact that high concentrations of KI precipitate Agl, and make it partially assume the crystallin form. In this in- stance the cause is a diminution of surface; in the other cases, above described, there is a strong probability of a diminution of surface, especially with the agglutination of bacteria. Hitherto it has been tacitly assumed that there is no affinity be- tween adsorbents and dissolved substance. This occurs, though only in a few exceptional cases. Thus B. S. G. Hedin *^ has shown that certain enzymes (trypsin and rennin) are irreversibly adsorbed from water, but they may be displaced by other substances (casein, serum, grape sugar). L. Michaelis demonstrated that acid kaolin could adsorb only basic or amphoteric dyes, while basic clay could adsorb only acid dyes. Similar experiments have been performed by various SURFACES 29 ..c other experimenters with wool, filter paper, etc. Since the chemical constitution of many of these substances is unknown and others can- not be prepared free of electrolytes, H. Bechhold considered it de- sirable to settle the problem by, using substances possessing definite constitution and which were easily obtained in a pure state. As such adsorbents, he chose naphthalin (CioHs, neutral), naphthol (C10H7OH, acid),naphthylamin(CioH7NH2, basic), amidonaphthol (C10H6OHNH2, amphoteric). These substances were freely suspended in water and shaken for several minutes with solutions of various dyes; they were filtered off and washed until the filtrate was practically colorless. The dye solutions employed were those generally used in microscopic technic. The results of my staining experiments are shown in the accompanying table. From these experiments it is evident that in most cases, even with the neutral naphthalin, there is at least a faint staining. The coloration is so slight, that it certainly could not be recognized in a microscopic specimen and it is evidently to be attrib- uted to mechanical adsorption. It may be seen at a glance how re- markably the staining differs, depending on the chemical constitution of the stained substance, neutral naphthalin is not deeply colored by any stain. Naphthylamin and amidonaphthol are always most strongly stained by the acid dyes, and naphthol and amidonaphthol by the color bases. Thus we see that the chemical constitution of the adsorbent plays a very important part in the distribution of the dissolved substance between the solvent and the dispersed phase. Berczeller and Czaki* reached analogous results with the adsorp- tion of alkaloids (cocain, atropin, etc.), by various powders (starches, coagulated albumin, which acted as weak acids and adsorbed most strongly, whereas alkaline CaCOs adsorbed least) . That we are deal- ing with electro che7nical phenomena in the above cases is still more evident when we observe how the addition of electrolytes affects adsorb- abihty. Wool, which is dyed particularly well by basic dyes in a neutral bath, takes them up still better from an alkahne bath, but it is also dyed in an acid bath with acid colors. Still better evidence lies in the fact that the cations of neutral salts increase the dyeing of acid colors in proportion to the valence of the cation (W. M. Bayliss). We must furthermore mention the fact that supplementary chemical reactions may occur between adsorbent and adsorbed substances, which may lead to a fixation that makes the process irreversible, i.e., a true chemical combination may result. The occurrence of this condition is characterized by the fact that it requires a certain length of time, whereas according to H. Freundlich, the adsorp- tion balance is established in a few minutes. Moreover, such a de- 30 COLLOIDS IN BIOLOGY AND MEDICINE 1 '^midonaphthol Naphthalin. /3-Naphthol. /3-Naphthylamin. (freshly precipi- tated). Basic dyes: Methylene blue very faint blue dark blue very faint blue LöfHer's blue ^ faint blue dark blue almost un- colored blue Carbol-fuchsin faint red reddish almost un- colored faint blue deep red Crystal violet bluish deep blue deep violet blue Bismark brown iiiTiit fi iTipn brownish practically unstained LlllOLcmiCVJ. Gram's Stain: Anilin water — gen- light violet deep violet faint violet deep violet tian violet Treatment with io- almost com- din-potassium io- pletely de- did solution. colorized dark blue blue dark blue Washing off with al- almost com- undeter- completely deep violet cohol. !- pletely de- mined be- decolor- (as far as colorized cause too soluble ized could be deter- mined in view of great solubil- ' ity) Acid dyes: Eosin unstained faintly yel- unstained very faintly pink reddish yel- red Aurantia yellow low yellow low Picric acid unstained almost un- stained very faint yellow almost un- stained Alizarin (dissolved in brownish brownish violet violet KOH) violet Gallein unstained almost un- stained almost un- stained faint brown- ish red Chrome violet almost un- almost un- almost un- red stained stained stained Amphoteric dyes: Benzopurpurin reddish reddish (somewhat deep er than the other s and mot- tled blue) reddish red Janus red reddish deep red reddish deep red Mixtures: Triacid faint bluish deep bluish very faint dark bluish green green violet green 1 Löffler's blue is a methylene blue solution with a trace of alkali; Carbol-fuchsin is a fuchsin solu- tion containing phenol. Both of these as well as Gram's stain are employed in staining bacteria. The latter affords an opportunity for the differentiation of certain bacteria and cocci. Although many bacteria are entirely or partly decolorized by subsequent treatment with the iodin-potassium iodid solution and washing with alcohol, many retain their stain. SURFACES 31 layed supplementary process may indicate a slow diffusion of the adsorbed substance into the adsorbent, as J. Davis has demon- strated in the case of the adsorption of iodin by charcoal. In the fixation of dyes by textile fibers we can assume the probable occur- rence of secondary chemical processes. I believe that many mis- understandings in the moot questions of toxin-antitoxin fixation might have been avoided, if there had been a clear understanding of the various phenomena which may occur in the course of an adsorption process. Finally, it must be mentioned that catalyzers are adsorbed (all organic ferments are colloids). By reason of adsorption, the reaction in a solution may be stopped; or in other cases, the reaction may be accelerated upon the adsorbent. Thus oxidations may be brought about by concentration of oxygen on the adsorbent or reductions by concentration of hydrogen (C. Paals' reduction of nitrobenzol by colloidal palladium and other chemical reactions.) The fundamental conception of adsorption is so illuminating that the attempt has been made to explain a large series of biological processes as adsorption phenomena (enzym action, union of toxin and antitoxin, etc.). I cannot better express the results of all this work than in the words of W. Biltz,*'* who^ in another connection, says, " The testing ... of the material in accordance with an exact method, such as is involved in the use of a formula, offers the worker a rather mingled pleasure, as may be noted from the great difference between the results of experiment and of calculation. If it were not for the novelty of the subject investigated, . . . the result which is so frequently accepted for the sake of the principle, would be of very little importance." I should like to add further: The adsorption formula is a rock which, Lorelei-like, has magically attracted numberless scientific voyagers only to wreck many of them. Every complicated phe- nomenon of higher organisms, which consists partly of chemical, partly of solution phenomena and perhaps partly of true adsorption, must assume the general character of an adsorption and have a formula which seems a cross between a chemical process and a solu- tion. Thus, if a biological phenomenon seems to suit the formula of an adsorption, this may be merely a sign post, which, alas, many an investigator may mistake for the goal. What then is the biological significance of what we have distinguished as chemical combination, solution and adsorption? Here we approach the most important principle governing the processes in the living organism; it is what P. Ehrlich describes as distribution. The de- veloping and the fully developed organism are constantly receiving 32 COLLOIDS IN BIOLOGY AND MEDICINE food materials, which are taken up at the places where needed, stored, and when necessary given up again. In other words, the organ- ism, plant as well as animal, is a vessel containing an aqueous solu- tion in which various colloids exist as dispersed phases. The balance which rules at each moment is disturbed by food material entering the vessel and by the metabolic products developing in it, and these become distributed between the solvent and the dispersed phase, the organ-colloids. In this entire chapter we have treated the simple case occurring when a single dispersed phase exists in a single dispersing medium (solvent). We may still assume without serious error, that there is one dispersing medium; but instead of one dispersed phase in the organism we have dozens, perhaps hun- dreds, of dispersed phases. Each individual class of cells is a dif- ferent dispersed phase with different properties. Only in this way can we understand how the assimilation products are sorted, stored up and changed into the tissues of the several organs. To quote a single example, we find (according to a table in E. Abderhalden's Textbook of Physiological Chemistry) that there are the following distributions: to 1000 parts by weight Serum of ox. Blood corpus- cles of ox. Sugar Cholesterin Lecithin Soda Potash Lime Ferric oxid 1.05 1.238 1.675 4.312 0.255 0.119 3^379 3.748 2.232 0.722 1^671 It is difficult to imagine a more unequal distribution. For in- stance, potassium and sodium salts enter the circulation as electro- lytes to the same extent; and there can be no question of any irreversible chemical combination of these salts either with the serum or with the blood corpuscles, since both diffuse away when the serum or blood corpuscles are brought in contact with pure water. There is thus a condition of equilibrium in the blood by which the hlood corpuscles dissolve or adsorb proportionately more potassium salts, while the serum albumin dissolves or adsorbs more sodium salts. This is not a unique case, for humus adsorbs chiefly the potassium salts from a mixture and permits the sodium salts to pass through. With iron the conditions are different; it certainly must enter the organism SURFACES 33 in some mobile form, yet at the places where blood corpuscles are formed it is chemically fixed as hemoglobin. We may say d priori that the products of dissimilation become very soluble in the dispers- ing medium, being dissolved or adsorbed but slightly by the dispersed phase; in fact they are not chemically fixed at all, so that they leave the body chiefly in the urine; they are, indeed, crystalloids of which only so much is retained in the blood by solution or adsorption as is necessary for a proper balance. We must here curtail our remarks and refer the reader to the chapter on Distribution of Substances and Metabolism. What applies to the substances necessary for the maintenance of the organism applies also to such foreign substances as have a toxic or pharmacodynamic effect. It is a principle that such foreign bodies as are chemically fixed, permanently injure the affected cell; narcosis seems to me, to be a typical example of simple solution, a process that is completely reversible. [Permanent injury may also be caused by the breaking of an emulsion. See M. H. Fischer and Marian O. Hooker, " Fats and Fatty Degeneration," John Wiley & Sons, New York, 1917. Tr.] Between these extreme cases there are substances which are adsorbed, and are active even in small doses, though larger doses do not cause materially greater damage. Under favorable conditions these processes may be reversible. The details are treated of in the chapter on Toxicology and Pharmacology. Finally, I wish to refer to the chapter on Immunity Reactions where it is an important question whether chemical combination, solution or adsorption obtains. Surface Skins. Absolutely pure water has no surface viscidity, which means that a metal or glass disc suspended by a thread at the surface performs as many oscillations after a single turn as it does when immersed. The slightest impurities may, however, suffice to cause a marked retarding effect at the surface. On page 25 we saw that substances which lower the surface tension of a fluid concentrate and spread out at the surface, so it is to be expected that the surface will have a different viscidity from the interior. Poggendorff and Plateau were the first to study the formation of skins on fluids, but we are indebted to M. V. Metcalf,* G. Nagel * and E. Rohde * for recent studies that have clarified the subject. It was shown as the result of these investigations that colloids and substances at the border line between colloids and crystalloids, especially many dyes, as fuchsin, methyl violet, peptones and several other substances, concentrate at 34 COLLOIDS IN BIOLOGY AND MEDICINE the surface of aqueous solutions and form a layer, which at first is easily movable. In a short time there is a change in this layer. In the case of staining solutions the surface appears dull after an hour and gradually there is formed a solid layer which histologists and bacteriologists know to their sorrow. It is, therefore, necessary to filter aqueous solutions of stains each time before using, even though they have been protected from dust. The dye concentrated at the surface undergoes chemical changes which the investigations showed to be independent of the gas upon the surface (it might have been attributed to oxidations with oxygen or to CO2, etc.). What has been said of dye solutions occurs also in the case of peptone solutions, as Metcalf demonstrated. The thickness of the layer which will just form a solid skin has been measured, and found to be, for peptone 3 fxß (Metcalf), for albu- min 3 to 7 jU;U (Devaux). Thus it is probably many times greater than the hypothetical diameter of a molecule, perhaps even equaling the radius of molecular attraction. The process of skin formation may be very much hastened by amplifying the surface, i.e., by shak- ing the fluid or passing gas through it. Thus W. Ramsden * was able to remove by shaking practically all the albumin from an albumin solution. The albumin passed into the foam and there formed solid skins. This phenomenon is of greatest practical impor- tance, since the solidity of meringue, whipped cream and beer foam evidently depends upon it. In the case of beer, the rising bubbles of CO2 carry foam-forming colloids at their surfaces and conversely the beer foam exerts a tension (pressure) which hinders the escape of CO2 and thus keeps the beer fresh for a longer time. Everyone who has worked with colloidal solutions knows how high the gas pressure must be in order that a stream of gas may be forced through a solu- tion which has once formed a layer of foam. The formation of a skin on boiling milk is evidently to be classed among these phe- nomena. The significance of this process for the coagulation of fibrin has been indicated on page 299. To these phenomena belongs 'Hnadivation of ferments hy shaking" described by E. Abderhalden and Guggenheim * and independ- ently by Signe and Sigval Schmidt-Nielsen.* The formation of surface skins is so sharply characteristic, that mixtures of substances, which diminish the surface tension of water to different degrees, may be separated by shaking. Until now, we have had only the qualitative investigations of W. Ramsden,* who determined the predominance of saponin in the foam on shaking a mixture of saponin and albumin. (Saponin lowers the surface tension of water more than albumin.) SURFACES 35 Hitherto we have only regarded the formation of surface skins at the interface fluid/gas. Such skins may be formed, however, at the interface fluid/fluid or fluid/solid, provided only that the sub- stance in question diminishes the surface tension of the water with reference to the other fluid or the solid phase. J. Ziegler has in- formed me (in a private communication) that on shaking benzol, toluol, etc., with water containing albumin or gelatin, the benzol or toluol forms above the water an emulsion which contains the colloid, and that with repeated shaking, the major part of the colloid may be removed from the aqueous solution. Shortly after this communication, there appeared a publication by Winkelblech * which not only confirmed these facts but called attention to the fact that through the formation of an emulsion mere traces of colloids could be detected. This phenomenon has long been recognized as a very disturbing factor by organic chemists. On shaking reaction mixtures with ether or benzol, such emulsions frequently form and are very difficult to separate. We know now that these emulsions are to be attributed to the formation of colloidal reaction products. H. Bechhold and J. Ziegler used the method of shaking out the foam for the separation of albumoses (Witte's peptone) into their components. They shook a 10 per cent aqueous solution of Witte's peptone with ether, separated the ether foam from the aqueous solution and again shook it out with ether. Then the ether was permitted to evaporate from the foam, and the residue was dissolved in ten times its volume of water, and this solution was again shaken out with ether. After thus treating the solution from three to five times, two substances were obtained, one of which re- mained in the water in clear solution and became turbid when treated with 24 to 25 per cent of ammonium sulphate. By this procedure a separation of two components was obtained, but it is a question whether the water-insoluble portion was present in the original solu- tions or was formed by the shaking, like Metcalf's peptone skins. The slight diminution in concentration is in favor of the first view. By "shaking out the foam " a separation of the slightly water-soluble hetero-albumoses and the remaining albumoses was accomplished. This spreading of colloids and the formation of colloid films at the interface between two fluids is a phenomenon very frequently observed. G. Quincke *^ showed that gum coflects at the interface between oil and a gum arable solution. Pharmaceutical emulsions accordingly consist of ' oil spheres surrounded by a film of gum. And when oil is emulsified with albumin, oil spheres surrounded by a film of albumin are formed (Ascherson *). I have attributed the so-called ''serum films surrounding the globules in milk" to the for- 36 COLLOIDS IN BIOLOGY AND MEDICINE mation of surface skins, see page 346. Where oil globules occur in aqueous solutions containing colloids, it may be assumed that they are surrounded by a film of colloid which prevents them from run- ning together and forming larger drops of fat. This is exemplified in the emulsion of fats in the intestine and in the milky turbidity of the serum after ingestion of fat as well as in the oleaginous and resinous emulsions in plants, e.g., in the milky sap of the Euphorbiace. As has been already stated the same conditions obtain for the interfaces between fluid/solid as for fluid/fluid. H, Bechhold*^ explains the action of protective colloids (see p. 11) as a mani- festation of this phenomenon. Protective colloids form colloidal films about the substance in suspension and thus impede the coalescence (flocculation) of the separate particles. Consequently surface pellicles afford stability to the metal sols, and permit their practical utilization. On the other hand, suspensions and hydrophobe colloids, depend- ing on circumstances and the surface tension, may either pass from one fluid into another fluid with which it is not miscible, or they may concentrate at the interface (Reinders). This must be considered in all studies on the distribution of colloids in the body, as in staining with colloidal dyes and the injection of colloidal metals and possibly even infection with micro-organisms. [According to the views of Martin H. Fischer and Marian 0. Hooker,^ we must distinguish between the making of an emulsion ^ See Martin H. Fischer and Marian O. Hooker, Fats and Fatty Degen- eration, John Wiley and Sons, New York, 1917, where references to their earher pubhcations may be found. " Both W. D. Bancroft and G. H. A. Clowes at the Urbana (1916) meeting of the American Chemical Society, in their discussion of our own views regarding the importance of colloid solvates (colloid hydrates) for the stabilization of emul- sions, found in om- views something irreconcilable with their notions of the im- portance of interfacial films and of surface tension changes in these. While we do not wish to insist upon a harmony where such may not be desired, there is, of course, nothing mutually exclusive in the ideas of solvation, of changes in surface tension, and — at times — the formation of a continuous third phase between the two chief substances making up an emulsion. When "water," according to our notion, becomes a "colloid hydrate," the properties of the second liquid are different from those of the first, and these properties include surface tension, viscosity and distribution between two phases. But, we repeat, these factors to which Plateau, Quincke and Pickering first directed attention are not by them- selves able to explain all the phenomena observed. Where Clowes holds that an emulsion of oil is stabilized through sodium oleate because the substance reduces the surface tension of water, we would say that stabilization has ensued because the oil has been divided into a highly hydratable sodium soap. When the addi- tion of calcium destroys this emulsion, it is not because of complicated changes in a surface film, but simply because calcium oleate is an only slightly hydratable SURFACES 37 .J and its stabilization. The making of an emulsion is essentially a mechanical process concerned with the mere obtaining of the sub- division of one liquid in a second, as oil in water. The problem of the stabilization, after such a subdivision has been brought about, is a totally different matter. In a certain sense the main feature in this stabihzation consists of the getting rid of the water as such in the emulsion and substituting for it a hydrated colloid. An emulsion is stabilized through any so-called emulsifying agent only because this emulsifying agent is a hydrophilic (lyophilic) col- loid. Oil, for example, cannot be permanently emulsified in water in amounts exceeding a fraction of one per cent, but in a medium in which the water is bound to an emulsifying agent as a hydrate (or solvate) the oil content can be carried to a very high figure (50 or 60 per cent) . When, for example, acacia is used as an emulsifying agent, it means that the permanent emulsion is made permanent because the acacia unites with the water to form an acacia-hydrate. After the stabilization of an emulsion has been accomplished through the production of a colloid hydrate, secondary concentration effects may be brought about which lead to a concentration of the colloid material upon or in the surface of the oil droplets but these secondary effects are not to be confused with the primary ones neces- sary for the stabilization of the emulsion. W. D. Bancroft asserts in a review of Fischer's book that there are no criteria which these alleged compounds (the solvates) could satisfy. J. of Ind. & Eng. Gh., vol. 9, No. 12, Dec, 1917. Bancroft observed that while soaps of mono-valent cations used as emulsifying agents for oil and water, promote the formation of emulsions like cream, in which oil is dispersed in a continuous water phase, soaps of di- and tri-valent cations form emulsions of the opposite type like butter, in which water is dispersed in oil. Ban- croft considers that soaps of sodium or potassium, being readily dis- persed in water but not in oil, form an interfacial film or membrane, the surface tension on the water side of which is much lower than on the oil, and that consequently an emulsion of oil in water is formed, soap. Free water, in consequence, appears in the mixture, and the oü separates out in gross form, as described above, for only very Httle oil can be permanently subdivided in "pure" or "free" water. We describe the consequences of such changes from highly hydratable to less hydratable soaps upon the stabüity of an emulsion on p. 49. Neither do we wish our statement that an agreement is possible between Clowes' and our views on simple emulsions to be expanded to include his behefs regarding the biological behavior of the fat in living cells. We long ago gave up the notion of lipoid membranes about cells and the complex notions of their changing permeability to which Clowes and many authors still hold." 38 COLLOIDS IN BIOLOGY AND MEDICINE 1 ^/HzO H20 ^4 HgO oilL biL Emulsifying Na oleate alone agent or Na oleate in or film excess of Ca oleate equiv Emulsion formed Oil in water Equiv. chem. proportion Na oleate and Ca oleate Ca oleate alone or Ca oleate in excess of Na oleate equiv. Critical point Water in oil Fig. 4a. Conversion of oil-water to water-oil emulsion. while soaps of calcium and magnesium, being readily dispersed in oil but not in water, form a film, the surface tension on the oil side of which is lower than on the water, and consequently an emulsion of water in oil is produced. Clowes showed that emulsions of oil in water could be converted into emulsions of water in oil and vice versa by varying the propor- tions of certain electrolytes added to the system. When equal volumes of oil contain- ing fatty acid and water con- taining NaOH were shaken together sodium soap was pro- duced and an emulsion of oil in water formed. On shaking this emulsion with increasing proportions of calcium chlor- id, a critical point at which oil and water separated into two distinct layers was ob- served when the CaCl2 was added in sufficient amount to convert half the sodium soap into cal- cium soap. Further addition of calcium chlorid led to the forma- tion of a stable emulsion of water dispersed in oil. Conversely, the latter emulsion could be reconverted through the critical point into one of oil in water by shaking with the requisite proportions of sodium soap or caustic soda. (See diagram. Fig. 4a.) Clowes attributes these transformations to variations in the sur- face tension relations of the water and oil phases, caused by varia- tions in the proportions of the hydrophilic sodium soaps which lowers the surface tension of the water phase, and the lipophilic calcium soap which lowers the surface tension of the oil phase. An emulsion of oil in water is produced when the surface tension is lower on the water than on the oil side of the stabiUzing film or membrane formed by concentration of the emulsifying agent at the oil-water interface. A critical point occurs when the surface tension is equal or compensa- tory on both sides of the film, and an emulsion of water in oil is formed when the surface tension is lower on the oil side than on the water side. Electrolytes appear to exert a marked effect on emulsion equilib- rium, those having a more reactive or more readily adsorbed anion appear to promote the formation of emulsions of oil in water, while those having a more reactive or readily adsorbed cation exert the reverse effect, promoting the formation of emulsions of water in oil. SURFACES 39 .J The antagonistic effects exerted by electrolytes of these two opposing groups appear to correspond sufficiently closely with those observed by OsTERHOUT in experiments on Hving cells to suggest the possibihty that variations in permeabihty exhibited by protoplasm under the influence of various salts might be attributable at least in part to reversible transformations of the marginal layer of protoplasmic material between systems in which a non-aqueous phase is dispersed in an aqueous, which would be relatively freely permeable to water, and the reverse type of system in which an aqueous phase is more or less surrounded by a non-aqueous film which would be impermeable or relatively less permeable to water. A Traube capillary pipette was employed to study the influence exerted by given salts individually, and in combination, on the rela- tive degree of dispersion of interfacial soap films in oil and water. Aqueous solutions containing caustic soda or soap and various con- centrations of the salts to be tested were allowed to flow from the capillary pipette through neutral oil or oil containing free fatty acid, and the number of drops produced served as an index of the dis- persing or protective effect exerted by the electrolytes in question on the interfacial soap film. Those electrolytes which possess a readily adsorbed anion appear to cause an increase in the number of drops, which corresponds with a lowering of the surface tension of the water phase, a destruction of the surface film, and an increased permeability of the system to water. Those electrolytes which possess a readily adsorbed cation exert the reverse effect, diminish- ing the number of drops, which indicates diminished dispersion or destruction of the film and a diminished permeability of the system to water. For example, a 0.001-m NaOH passed through ohve oil gave 44 drops; the addition of NaCl to a concentration of 0.15-?7i raised the number of drops to 300; the addition of CaCl2 at a concen- tration of 0.0015-w lowered the number of drops to 24; while a system in which O.OOl-w NaOH was employed in conjunction with O.lS^m, NaCl and 0.0015-w CaCl2 gave 44 drops, corresponding with the original system and indicating that under the conditions of the experiment NaCl and CaCl2 exert an antagonistic or compensa- tory effect upon one another in the molecular ratios of 100 : 1. In similar experiments with other electrolytes, anesthetics, etc., the ratios in which antagonistic effects were produced corresponded sufficiently with those in which the substances in question exerted antagonistic effects on marine and other organisms as to suggest the possibility that these physical systems may afford a crude model of the mechanism underlying the control of permeability in protoplasm. Salts of magnesium and other substances which exhibit abnormalities 40 COLLOIDS IN BIOLOGY AND MEDICINE in biological systems exerting under varying conditions a protective or destructive effect on the protoplasmic film exhibit similar abnor- malities in emulsion and drop systems. Magnesium salts function as protective agents hke calcium salts when added to a soap solution which is passed through oil; but as destructive agents like NaOH, NaCl or KCl when added to a dilute solution of NaOH which is passed through an oil containing fatty acid. Closely parallel results were observed between the drop system de- scribed above, the process of blood coagulation, the lethal dose of given electrolytes in mice when injected intravenously, the hemolysis of blood corpuscles by complement and amboceptor, etc., a common critical point being observed in these widely diversified systems. Clowes considers that these experiments lend substantial support to to the view that while protoplasm as a whole consists of a system approximating more nearly to a dispersion of the non-aqueous phase in the aqueous, the extreme marginal layer of protoplasm may be looked upon as an emulsion or gel-hke system consisting of two con- tinuous phases in which fluctuations in permeability to water and water-borne substances may be caused by variations in the propor- tions of metabolic products, electrolytes, etc., a slight change in the system in the direction of water surrounded by the non-aqueous phase leading to a diminution in permeability, while a change in the reverse direction, towards a system in which the non-aqueous phase is more completely dispersed in the aqueous, would lead to an increased permeability. Substantial support is lent to this point of view by Osterhout's observations that the conductivity of Laminaria tissue is raised by exposure to a solution of NaCl, lowered by CaCl2, but unchanged when exposed to a mixture containing 100 molecules of NaCl and one of CaCl2. Life can only be maintained within certain ranges of electrical resistance or permeability and an increase or decrease in permeability beyond given limiting points is no longer reversible and invariably causes death. Wilder D. Bancroft: Jour. Phys. Chem., 17, 501 (1913). G. H. A. Clowes: Proc. Physiological Sec- tion, International Medical Congress, pp. 105-114, London, 1913. Proc. Soc. Exp. Biology and Medicine, 11, pp. 1-3, 4-5, 6-8, 8-10 (1913). Jour. Physical Chemistry, 20, p. 407 (1916). Proc. Soc. Exp. Biology and Medicine, 13, pp. 114-118 (1916). Science, 43, pp. 750-757 (1916). Tr.] CHAPTER HL SIZE OF PARTICLES, MOLECULAR WEIGHT, OSMOTIC PRESSURE, CONDUCTIVITY. For the chemist wishing to discover the constitution of a chemical substance, the determination of the molecular weight is of great im- portance. Much time has thus been spent on determining the molecular weights of the bio-colloids, such as albumin, starch, hemoglobin, etc., and in the following pages we shall show what probability for success attends these efforts. A soluble substance, placed in a suitable medium which produces no chemical change, distributes itself uniformly. In the case of crystalloids, it is impossible by either optical or mechanical means to recognize the particles into which it splits up.^ We shall see that crystalloids are frequently broken up into their molecules. Many colloids are soluble also. If we examine their solutions in the ultra- microscope which permits a hundred thousand fold magnification,^ we can recognize numerous particles. In the case of artificial colloids (gold and silver hydrosols) in which we are certain that all the dis- solved particles are visible, according to R. Zsigmondy we are in a pos- ition, as will be shown by the following considerations, to calculate the approximate weight of each individual particle. Let 1 gm. of colloid be dissolved in 1 liter of water, then every 1 cubic millimeter contains 1/1000 milligram of colloid. If by counting under the ultra- microscope, we determine that each cubic millimeter contains 1000 particles, we know that each particle weighs one millionth of a mil- ligram. We can easily calculate the diameter of a particle by supplying the specific gravity and assuming that each particle is a sphere. However, as soon as we become uncertain whether all the particles are visible, which is the case with most hio-colloids, the opti- cal method fails us. Under these circumstances, we can determine the size of the particles by ultrafiltration. If we sift grains, we know that those which pass through are smaller, and those which are 1 [Optical inhomogeneity of sugar solution has been demonstrated. Van Calcar and L. de Bruyn separated sodium sulphate from solution by high speed centrif- ugation. Tr.] 2 [The ultramicroscope makes visible otherwise invisible particles but does not magnify beyond the power of its component compound microscope. Tr.] 41 42 COLLOIDS IN BIOLOGY AND MEDICINE held back are larger than the meshes of the sieve. If we know the size of the meshes and have several sieves with meshes of different size, it is easy to determine the average size of the grains by letting them pass through the different sieves. H. Bechhold's determina- tion of the size of the particles depends on this principle. Ultrafilters (jelly filters) with different sized pores serve as sieves (see pp. 99 et seq.). Since there are several methods for measuring the size of the pores (see p. 100), it is possible to determine definite limits for the size of the colloid particles.^ Are the particles thus found identical with molecules? In the case of metal sols we can immediately say, no. We know the molecular weight of metals and understand from it that there is no prospect of directly seeing the molecules of the elements with our present instruments. According to E. Riecke gold particles of 1 iJifj. diameter have a molecular weight of 300,000, but the molec- ular weight of gold is probably only 197, and the smallest particles we can see have a diameter of 5 ijljjl. It follows, therefore, that every recognizable ultramicroscopic particle consists of thousands of molecules. What are the facts in the case of particles whose size is determinable hy ultrafiltration? Since albumin, starch, etc., have unusually large molecules, it is probable that in them the molecule and particle size, as determined by ultrafiltration, are identical. This is all the more likely since these bio-colloids, like crystalloids, are distributed by means of the action of the solvent, whereas the metal hydrosols are brought into such minute divisions only by artificial means. But what is a molecule? It is the smallest portion of a compound or of an element that may exist alone. If we split a molecule of common salt we no longer have a molecule of NaCl but an atom of Na and an atom of CI. If we divide an albumin molecule, we still have complicated atom complexes but we have albumin no longer. Molecular weight is the weight of a molecule compared to that of an atom of hydrogen which equals unity. Consequently we are measuring not absolute, but relative sizes. The molecular weight is determined by purely chemical means. If, for example, we find in sodium benzoate, that there are 7 atoms of carbon (7 X 12 = 84), 5 atoms of hydrogen (5 X 1 = 5), 2 atoms of oxygen (2 X 16 = 32) and one atom of sodium (1 X 23 = 23), we should know that the molecular weight must be at least 144, because half atoms do not exist. The molecular weight might in fact be two or three times as large, which would have to be determined by other 1 [J. Alexander has recently proposed measurement of particle size by high speed centrifugation, " ultracentrifugation." Tr.] SIZE OF PARTICLES 43 chemical investigations and determinations on other chemical compounds of benzoic acid. From similar considerations the minimum value for the molecular weight of certain albumins are obtained. If a protein contains one per cent of sulphur, then its molecular weight must be 3200 times heavier than that of hydrogen. (The atomic weight of S = 32.) But there is every reason to believe that there are at least two atoms of sulphur in egg albumen, because one-half of the sulphur is easily split off, whereas the other splits off with difficulty. Thus egg albumen, with one and three-tenths per cent sulphur shows a molec- ular weight of 4900, and oxyhemoglobin a molecular weight of 14,800. Oxyhemoglobin contains 0.4 to 0.5 per cent of iron; pro- vided it contains but one atom of iron, its molecular weight mufet be 11,200 to 14,000. The figures obtained approach one another very closely. Another method of obtaining the molecular weight is based on Avogadro's law. This law says: "At the same temperature and equal pressure, different gases contain the same number of molecules per liter." Thus from the weight of a gas or of a vaporized sub- stance, the molecular weight can be determined, if we compare its weight with that of an equal volume of hydrogen gas. Avogadro's law was generalized by J. H. van't Hoff and extended to solutions. According to this generalization the "osmotic pressure" of a dis- solved substance is proportionate to the number of the dissolved molecules and is as large as if the substance were vaporized. If a sugar solution is placed in a porous clay cell which is so dense that water but not sugar may pass in and out,^ the sugar tries to expand like a gas and, as a result, water enters the cell and the solution rises in it. If a vertical tube has been attached to the cell, the osmotic pressure of the solution may be measured directly from the height to which the fluid rises. There are indirect methods also, the under- lying principles of which we cannot discuss here. They depend on the fact that in proportion to the osmotic pressure the boiling point is raised and the freezing point lowered. In ideal cases these changes are strictly proportionate to the concentration of the dissolved sub- stance in just the same way that the original volume of an ideal gas is reduced to one-half by double the pressure and to one-third by three times the pressure. Consequently by determining the freezing or boiling point of a solution, molecular weight may be de- termined. In the case of crystalloids this method is preferred to the direct reading of the osmotic pressure for the following reasons: It is exceedingly difficult to prepare a cell that really holds back crystal- ^ Such a chamber is said to be semipermeable. 44 COLLOIDS IN BIOLOGY AND MEDICINE loids, and, on this account, the danger of considerable error is always present. Moreover, in solutions with ordinary molecular weights, the osmotic pressures are so great that very bulky apparatus would be required. Thus, for example, the osmotic pressure of a 1 per cent aqueous solution of sugar at 15.5° C. is actually 0.685 atmosphere. The difficulties in the direct measurement of osmotic pressure of crystalloids do not exist however, in the case of colloids. Almost any membrane keeps back colloids and the small rises are easily measured. In order to remove the sources of error due to the pos- sible presence of crystalloids, we employ membranes which are per- meable for crystalloids instead of semipermeable ones (collodion sacs, animal membranes). These physical methods for determining the molecular weight rest on the assumption that the substance in solution is really broken up into molecules (colloids cannot be vaporized). This condition does not always exist in the case of crystalloids and only exceptionally with colloids. With crystalloids these methods yield figures that are either too low or too high. The former will occur if the sub- stance is incompletely split up — if two, three or more molecules continue to be united in solution. Under these circumstances we ob- tain one-half, one-third, etc., the osmotic pressure that a molecular subdivision would show. The ultramicroscope and ultrafiltration reveal in many solutions of bio-colloids particles of such a size as show no molecular subdivision by other methods; we may assume therefore that colloidal solutions for the most part contain molecular groups, and that there is no prospect of determining the true molec- ular weight by osmotic methods. The osmotic method yields a deceptively low molecular weight if, for example, the substance is dissociated further than into mole- cules. This occurs in the case of electrolytes. A very dilute NaCl solution that has dissociated into Na and CI ions shows twice the true osmotic pressure, so that the molecular weight might be set at half its real value. In this respect we may also make mistakes with colloids whenever they are ionized. The osmotic method shows us only into how many fragments a molecular complex breaks up in the par- ticular solution. It may give either minimum or maximum values for the molecular weight. Even in the case of crystalloids, the method must be employed with due consideration of all the conditions in- volved; for colloids it may be exceedingly deceptive. We know at the outset, because of the enormous molecular weight of colloids, that the lowering of the freezing point and the elevation of the boiling point must be very small indeed, requiring most delicate measure- ments. The matter becomes still further complicated by the fact SIZE OF PARTICLES 45 that crystalloids are adsorbed by colloids and cannot be completely removed by dialysis. Each crystalloid molecule or each crystalloid ion may thus falsely represent the osmotic pressure of a colloid mole- cule having perhaps a thousand times its mass. The coefficient of diffusion may be employed in the determination of the molecular weight of crystalloids, but in the case of colloids it gives information concerning only the average size of the particles. The method is not much impaired by the increase in the size of the molecule, because it is only the square of the coefficient of diffusion which diminishes proportionately to this increase. The adsorption of crystalloids, on the contrary, is also in this case a source of error because every crystalloid molecule or ion acts as a team-mate of its colloid particle and hastens its rate of diffusion. The objection to the principles governing the calculation are mentioned on page 53. Before we come to concrete examples we shall mention one other method which may enlighten us concerning the particle content of a solution — the conductivity. In a solution the electric current is carried only by the electrically charged particles (ions). In an NaCl solution this is done by the Na and CI ions; in a Na2S04 solution, 2 Na ions and 1 SO4 ion, that is 3 ions, take part. The assimiption is that many molecules are completely or almost completely split into ions; this actually occurs in the case of strong electrolytes when in great dilution. The conductivity thus affords us fractions of the molecular weight: minimum figures (values less than actual). My chief purpose in making these statements is to show what facts may be deduced from the various methods used in determining the molecular weight; they give only limiting values, so that no con- clusion is to be drawn from any one method. The following remarks will show what difficulties stand in the way when we try to learn the size of the colloid molecule. Among the colloids whose chemical composition is best known are the soaps. As was found by F. Krafft and A. Smits, very dilute soap solu- tions showed a well-marked rise in boiling point; but this did not rise in proportion to the concentration of the soap, as may be seen from the following table by A. Smits for sodium palmitate: Concentration in mols. Rise in boiling point, °C. 0.0282 0.1128 0.2941 0.5721 0.024 0.045 0.050 0.060 46 COLLOIDS IN BIOLOGY AND MEDICINE Though the concentration is increased twenty fold, the boihng point rises only two and one-half times. In a solution of 19,5 per cent sodium stearate, F. Krafft found absolutely no rise in the boiling point as compared to pure water. Let us examine the conclusion of W. Biltz and A. v. Vegesack based on their critical study of the osmotic method. True colloids, like iron oxid and tungstic acid show a small osmotic pressure, only so long as they contain electrolytes. As the electrolyte vanishes, the colloid particles aggregate to larger complexes which then cease to show any osmotic pressure. For the existence of these colloids, some electrolyte content is an absolute necessity. When these investigators studied '' colloid electrolytes," particularly colloid color salts (congo red, night blue and benzopurpurin) whose constitution, molecular weight, etc., were determined by chemical methods, they obtained results which especially in the case of congo red must be closely examined. Congo red has the formula C32H22N6- S206Na2, and being a sodium disulphonate, is a strong electrolyte. Its molecular weight (M) = 696. On account of its electrolytic dis- sociation into 3 ions (2 crystalloid and 1 colloid) we would expect an osmotic pressure three times as much as its molecular weight would indicate. Instead of this W. M. Bayliss and also W. Biltz and A. v. Vegesack as well as Donnan and Harris obtained by dialysis against pure water a pressure which was approximately 5 per cent lower than would have been obtained had the undissociated molecule been active. The explanation is not difficult. Let us desig- nate by R, the acidic color radical of congo red, then congo red has the formula Il.Na2. In solution a portion becomes ionized into R and Na Na, of which some, even though possibly only a small fraction forms with the H and OH ions of the water RH (color acid) and NaOH. This occurrence would be without much influence in chang- ing osmotic pressure if the measurement was made in a closed vessel in which the equilibrium was undisturbed. As a matter of fact the measurement is made in a membrane permeable for electrolytes. Consequently the NaOH which has been formed diffuses away and some fresh color acid (RH) may be formed. This process continues until practically only color acid remains in the membrane. Con- sequently in this instance we have not measured the high osmotic pressure of the electrolytically strongly dissociated color salt but that of the practically undissociated color acid. If measured against outer water containing electrolytes, it yields a very much lower osmotic pressure, equivalent to a value for M of 2088. We shall thoroughly understand this occurrence when we have become more familiar with the equilibria of membranes. See page 59. SIZE OF PARTICLES 47 We shall only indicate here, that when the colloid electrolytes (thus BiLTZ designates salts in which one ion is a colloid) and the electrolyte in the outer water have originally the same osmotic pressure, there is a gradual penetration of the outer electrolytes but the colloid electrolytes cannot escape. Consequently the osmotic pressure in the cell is the resultant of osmotic pressure of the colloid electrolytes plus that of the electrolytes which have entered. If the latter have an ion in common with the colloid electrolytes, we do not find, as might have been expected, that there is an equal division of the true electrolytes {e.g., NaCl), but on the contrary, outside the osmometer there is proportionately more NaCl the more dilute the NaCl solution is (see page 62) . The osmotic pressure of the colloid electrolytes is consequently depressed. From this it follows that the values determined by the direct osmotic methods require revision. E. H. Starling obtained 4 mm, Hg osmotic pressure for serum colloid in 1 per cent serum, thus ex- pressing an apparent molecular weight of about 50,000. E. W. Reid obtained a pressure of 369 mm. Hg for a 1 per cent hemoglobin solu- tion from which is deduced an apparent molecular weight of about 65,000, a figure which approaches the values obtained by the diffusion method by R. 0. Herzog and Sv. Arrhenius. These figures are 4 to 10 times greater than those determined for the molecular weight by chemical means. As a matter of fact the theory of the direct measurement of osmotic pressiu-e is so difiiciilt that I do not know any results which are not sus- ceptible of adverse criticism (see F. G. Donnan's theory). As the re- sult of direct measurement, we know that colloid solutions actually have an osmotic pressure and that it increases with the amount of dispersion. Theoretical considerations, however, show us that this osmotic pressure must be low. Theoretically, all solutions which contain the same number of particles of the dissolved substance exert the same osmotic pressure. Thus all normal solutions (leaving out of consideration dissociations, associations and other changes) exert the same osmotic pressure, namely, 22.4 atmospheres. Normal solutions are such as contain the same number of molecules, namely, one gram molecule per liter. A normal salt solution is one containing 58.5 gm. NaCl per liter and a normal hydrogen solution is one containing 2 gm. of hydrogen per liter (theoretically). By various means which yield rather concordant results it has been determined that 2 gm. H contain 6.1 X 10"^ molecules^ or fragments of molecules or molecular com- plexes, i.e., particles in solution exerting 22.4 atmospheres of osmotic ^ This figure (6.1 X 10-') is called Avogadro's figure, and the various methods for deriving it give quite uniform values. 48 COLLOIDS IN BIOLOGY AND MEDICINE pressure. When we consider that the finest gold particles of a col- loidal gold solution have a diameter of only 2/j,fjL^ (a dimension one- fifth that of ultramicroscopic visibility); according to Svedberg in order to make a normal colloidal gold solution we should have to cram 50 kilos of gold into a liter, which would have to contain 6.1 X 10^^ such gold particles. It is naturally impossible to do this; the most that can be dissolved, experimentally, is one gm. per liter. Under favorable conditions such solutions have an osmotic pressure of 4.5 X 10~* atmosphere, i.e., they would rise 4.65 mm. in an osmom- eter, or be in equilibrium with electrolytes which modify the electro- lytic or hydrolytic dissociation of colloid electrolytes. It is unnecessary to catalogue all the dozens of fruitless investi- gations of the molecular weight of colloids by physical methods. They either gave surprisingly low molecular weights, when it could be shown on testing that the colloid was contaminated with crystal- loids, or the values were so small (the molecular weight so large) that they fell within the limits of error of observation, which means that it became doubtful whether the colloid studied had any osmotic pressure at all. The molecular weight is the expression of a chemical point of view which cannot be determined by ^physical methods for colloids. What we obtain by these methods are more or less nu- merous groups of molecules, usually in adsorption balance with irre- movable traces of crystalloids which cannot be separated, or which are in equilibrium with electrolytes that influence the electrolytic and hydrolytic dissociation of the colloid electrolytes. In the present state of the science we may only strive to determine the size of the particles in solutions of colloids having various equilibria. 1 [According to Zsigmondy particles 5mm in size may be seen with the aid of strong sunlight. — Tr.] CHAPTER IV. PHENOMENA OF MOTION. Brownian-Zsigmondy Movement. Upon examining a drop of milk under the microscope, we are at once struck by the appearance of the fat droplets on account of their strong refraction (a dark ring with a brilliant center). It is seen that they exhibit a certain oscillation (trembling). This chu,racter- istic oscillating movement is more intense with the smaller droplets (Fig. 5), whereas those having a diameter of more than 4 (j. do not show it at all. The phenomenon is named after the English botan- FiG. 5. Brownian movement of milk globules. (From O. Lehman.) ist, Robert Brown, who discovered it as early as 1827, in an aqueous suspension of plant pollen. It may be observed in every suspension or emulsion which is sufficiently fine. Particles of 1 ju diameter show a radius of translation of 1 ^t. The ''dance of the motes," the rush- ing hither and thither of the bright particles observed in the ultra- microscope,^ is nothing else than an enormously exaggerated Brownian movement, due to the fact that the particles are much smaller than those that may be seen under the microscope. Particles of 10 to 50 /jlh have a speed of more than 100 /x per second. These movements ^ R. Lorenz correctly calls attention to the fact that the great advance in our science does not date from Brown who observed the "oscillations" of microscopic particles but from R. Zsigmondy who recognized that particles of molecular dimensions are in a similar mobile state. (Frankforter Ztg. 4.6.11, I Morgenbl.) 49 50 COLLOIDS IN BIOLOGY AND MEDICINE seen under the ultramicroscope are comparable to the dance of the molecules in accordance with the Kinetic Theory of Gases. The speed of the particles is dependent on the viscidity of the dispersing medium and increases with a rise in temperature. It is very pertinent to enquire at this point whether we here see the move- ments of the molecules themselves. In a certain sense, this may be answered in the affirmative. Though we cannot yet say that this movement is inherent in the particles, i.e., that it would be carried out by the particles themselves, we may assert that it is caused by blows from the molecules of the solvent. A. Einstein and M. von Smoluchowski have independently de- duced from the Kinetic Theory of Gases, laws for the Brownian movement (extent of movement, influence of temperature and vis- cosity). It might be assumed a priori, that a particle floating in a fluid would remain at rest, for it simultaneously receives from all sides an equal number of impacts from molecules. The fallacy of this assumption is shown by M. von Smoluchowski in a very pretty comparison. If we play roulette for a long time the chances for gaining and losing are equal (disregarding the banker). If we play only a short time we win one day and lose the next. In other words the law of probabilities shows that the excess of molecular impacts which reach a particle in a given quarter suffice to give it movement one direction or another direction. The smaller the particle, the greater is the probability that the impacts will not arrest it and the stronger is its movement. The formula of von Smoluchowski as well as that of A. Einstein demands that -~ be constant for equal sized particles; A = amplitude, tq = viscosity, T = oscillation time. Th. Svedberg, by brilliantly devised methods, measured these values on colloid metals, in various dispersing media, and estab- lished the constants. It is true that the absolute figures Ä^the. measured and for the calculated amplitudes do not exactly ngree, but they are of the same order of magnitude; i.e., the values found are on the average three times as large as those calculated. Seddig, also, confirmed the quantitative increase of amplitude accompany- ing a rise in temperature. This is a remarkable agreement between the movements of small particles seen with the eyes and the hypothesis of the movements of gas molecules based on scientific imagination, which Kroenig in 1856 and Clausius in 1857 formulated mathematically (kinetic theory of gases).- All investigations that have since been undertaken PHENOMENA OF MOTION 51 concerning the laws of gases and the movement of colloidal particles have essentially agreed in showing that the laws of gases proved applicable to very dilute solutions of hydrophobe colloids and con- versely, that the laws of gases could be developed from the move- ments of colloid particles. Boyle's law asserts that the volume (v) of a gas is inversely proportional to the pressure (p) exerted upon it: V : v' = p' : p. According to Gay-Lussac's law the volume change of a gas having the temperature (t) is v = Vo {I + at) in which Vo is the volume at 0° and a is the coefficient of expansion. From the standpoint of molecular kinetics under doubled pressure twice as many moving particles are present in the same space. With increase of temperature (assuming the same pressure) at times fewer parti- cles are present than in the same gas volume at 0°. This assumes average values, though in fact, the number of particles in a definite volume varies from m.oment to moment. If this assumption is correct the average of the "instant values" must give values which sat- isfy Boyle-Gay-Lussac's law. M. von Smoluchowski developed mathemartically the relation between this law and the " instant values." He obtained experimental verification when Th. Svedberg counted the number of particles for an "instant value" directly in the ultramicroscope and R. Lorenz counted the particles in cine- metagraphs of the ultramicroscopic field. The assumption also bridges the gap between Thermodynamics, which studies phenomena on the basis of the involved energy and its transformations, and the Kinetic Molecular Theory, which views matter as the smallest possible particles in motion. The impact that our ultramicroscopically visible particles exert against the walls of a vessel is the pressure they exert, and it is measurable for a molecularly dispersed system as the osmotic pressure. The osmotic pressure, a, function of the mass and motion of a sus- pension whose particles are visible and measurable, was shown by ^►:mRiN to coincide with the requirements of the Kinetic Theory of Cases and of Thermodynamics; that is, with its energy content in the form of heat. The following considerations make this clear. . Under the influence of gravity the lower layers of the atmosphere have a greater density than the upper; i.e., the number of gas particles (molecules) in 1 cc. is greater in the immediate vicinity of the earth than at higher altitudes. This applies not only for gases but also for solutions or suspensions. " J. Perrin prepared a very fine suspension of gamboge which he placed in a tall cylinder. Gradually under the influence of gravity an equilibrium developed in which there was dense suspension at the bottom of the cyUnder with gradually diminishing concentration of particles in the upper layers — an atmosphere in 52 COLLOIDS IN BIOLOGY AND MEDICINE miniature (Fig. 6). By means of the ultramicroscope he counted particles in each layer of 0. 12 mm. The (osmotic) pressure of a particle may be calculated by the following formula applicable to the kinetics of gases: In no = lmgh{l -l^j; r. no and n are the number of particles counted in the unit volume at the levels 0 and h, m = mass, g = acceleration due to gravity, s = density of particles. The value of k proved to be 43 • lO-i'^. If this pressure of a single particle expresses the pressure of a single molecule (in solution or as a gas), the equation k = — appUes. Here N is the number of the molecules present in a gram molecule as determined by other methods, that is, 6 • 1023, T = 295°, A; is 43 • IQ-^K From this we derive the value ß = 2.1 cal. instead of 1.98 cal. From this the molecular weight of the gamboge particles was proved to be 3 • 10'. Diffusion. If pure water is layered over a concen- trated sugar or salt solution so that there is no mixing, it is found that after a certain time (hours or days), the sugar or salt passes into the water; we say that it diffuses into the water. If one chooses a colored salt solution, e.gr., copper sulphate, the path of the diffusion may be easily observed by the coloration. It is in prin- ciple the same process as when one per- mits a compressed gas to stream into the air — the distinction lies only in the differ- ence in speed. It has been shown that different sub- stances have very different rates of diffu- sion which are characteristic for these substances. These characteristic con- stants are called coefficients of diffusion. The coefficient of diffusion expresses the amount of substance which passes through an area 1 cm.^ per second ^ from a solution contain- ing 1 part per cc. ^ Because this time is so brief it is usually necessary to multiply the coefficients . by a large factor or to choose the day as the unit of time. c*.- I* ' ,.,'f.,. Pig. 6. Mastic suspension showing the effect of grav- ity. (Perrin.) PHENOMENA OF MOTION 53 .J It is evident that these values are free from any hypothetical con- siderations. It has, however, been shown that the coefficient of diffusion for crystalloids bears a certain relation to the molecular weight. Small molecules diffuse rapidly, large ones slowly. When suitable formulas are used there is very satisfactory correspondence, provided the molecular weights are not smaller than 50 nor larger than 500. A further advance was made by seeking to calculate from the coefficient of diffusion the molecular weights of colloids whose M was unknown. The results were not concordant because the moving units in colloidal solutions are not "molecules" but "particles," that is, complexes of molecules. If we connect the fact that the coefficient of diffusion decreases with increase in the size of the molecule, with what we know about the Browfiian-Zsigmondy movement, the relationship is surprising. We have seen that the movement is smaller as the particles grow larger and it is evident that, when we layer water over a metal hydro- sol, the strong translatory movements which we observe under the ultramicroscope must carry the hydrosol into the pure water. In coarser suspensions possessing only vibratory movements, we do not expect diffusion to occur.^ Svedberg measured the diffusion coefficient in different solutions of colloidal gold and calculated the size of the particles from the very simple relation (particle size in- versely proportional to diffusion coefficient). The experiments were carried out with two gold solutions which contained particles of 1-3mm and of 20-30ßiJ,, directly measured ultramicroscopically. There was a relatively good agreement between the results as cal- culated and determined. A direct relationship between the Brown- ian-Zsigmondy movement and the coefficient of diffusion cannot be experimentally established by methods free from criticism, because the hydrophobe hydrosols (e.g., colloidal gold, platinum and the like) which may be counted under the ultramicroscope cannot be prepared entirely free of crystalloids. Since every crystalloid molecule which is attached to a colloid particle must greatly accelerate the diffusion of the latter, we are confronted with a source of error that is uncon- trollable. A series of coefficients of diffusion have also been measured for hydrophile colloids, which though they have not the exactitude possessed by those of crystalloids, reveal a remarkable constancy so that they may be considered characteristic for the substance under consideration (Sv. Arrhenius, R. O. Herzog, Euler, Öholm) . ^ For the mathematical relations between diffusion coefficient, molecular weight and molecule or particle diameter see R. O. Herzog and L. W. Oholm. 54 COLLOIDS IN BIOLOGY AND MEDICINE This may be illustrated by the following figures: Diffusion coefficient D at 20° C. Molecular weight Particle diameter in MM. Substance. otherwise determined. calculated from D. Observer. Urea 110 73 66 38 14 10.5 7 7 6.6 5.9 3.6 3.3 60 92 110 342 j 973? I 2612? 40 91 111 337 2,430 4,440 10,000 10,000 11,200 14,200 37,700 44,900 0.34 0.51 0.57 0.98 2.65 3.57 5.36 4.60 4.88 5.46 8.88 9.76 Öholm Glycerin Resorcin a Cane sugar Graham- Inulin Stephan Öholm Dextrin Soluble starch (( Pepsin Rennin Egg albumen Emulsin Herzog Invertin a All the facts mentioned above indicate what Einstein emphasized even in 1905, that there is a gradual transition from crystalloid to colloid solutions, and that the translatory movements of the particles of a colloid correspond to the diffusion of crystalloid molecules. [Sir William Ramsay, in a paper entitled " Pedetic Motion in Relation to Colloidal Solutions" published in "Chemical News," Vol. 65, p. 90 [1892], stated as follows: " I am disposed to conclude that solution is nothing but subdivi- sion and admixture, owing to attractions between solvent and dis- solved substance accompanied by pedetic motion; that the true osmotic pressure has, probably, never been measured; and that a continuous passage can be traced between visible particles in suspen- sion, and matter in solution; that, in the words of the old adage, Natura nihil fit per saltum." Tr.] Diffusion in Jellies. Hitherto we have considered only diffusion in pure aqueous solu- tion; in the organism, however, it occurs in a more or less dense colloidal medium. When the concentration of the colloid is not very great, the diffusion is not much impeded. Until recently it was even believed that diffusion of a crystalloid solution in a jelly, e.g., gelatin or agar, occurred just as rapidly as in pure water. This was the result of employing unsuitable experimental methods. The investigations of H. Bechhold and J. Ziegler,*^ Kurt Meyer,* Peter Nell* and L. W. Öholm showed definitely that electrolytes and non-electrolytes experience a resistance in jellies which reduced the rate of diffusion by obstructing their paths, and that the inter- ference increased if the gel became more concentrated. PHENOMENA OF MOTION 55 Even the age of the jelly may have an influence. Thus, F. Stof- fel * showed (from H. Zangger's laboratory) that the diffusion- path of crystalloids in gelatin which was rapidly solidified is greater than it is in gelatin which was slowly solidified, but that this becomes equalized after several days. The rate of diffusion may be delayed or hastened through the presence of a third substance. This affects the diffusion in jellies to a much greater extent than in liquids. On page 69 et seq., we shall see that chlorin, iodin, nitrate, and other ions, urea, etc., favor swell- ing; on the other hand, sulphate, citrate, and other ions as well as alcohol, sugar, etc., as compared with pure water, diminish swelling, so that to a certain extent the meshes of the colloid network may be opened or closed. It is easy to understand that diffusion will occur less rapidly through narrow meshes than through wide ones. That such an influence on diffusion actually occurs was experimen- tally shown by H. Bechhold and J. Ziegler. They showed that the permeability of gelatin and agar jellies for electrolytes and non- electrolytes was increased by urea, whereas it was diminished by sodium sulphate, grape sugar, glycerin and alcohol. An increase is also produced by sulpho-groups, according to Böhi.* It is evident that every substance which increases the permeability of other substances, paves the way for its own passage as weh. If a jelly has been saturated with urea, the later coming particles of urea will diffuse more rapidly; conversely, sodium sulphate and grape sugar particles obstruct by their influence on the gel the pas- sage of the subsequent particles. There are other ways in which diffusion in a gel may be dis- tinguished from that in aqueous solution. We know from Chapter II that colloids adsorb other substances to a greater or less extent. By this means diffusion may be more or less impeded and under certain circumstances even entirely arrested. This may be ob- served with ease in the diffusion of dyes. H. Bechhold and J. Ziegler*^ showed that gelatin was deeply stained with methylene blue and thus the diffusion in the gelatin was impeded;* whereas the juice of red beets is a dye which is not noticeably adsorbed. Finally, if we observe that adsorption is strongly influenced by the presence of salts and non-electrolytes, and that an effect on diffusion is thus exerted, we shall see what great complications may appear when diffusion occurs in colloid media. Though a diffusion of colloids in aqueous media was long doubted the diffusion of colloids in jellies was positively denied. Thomas Graham held it to be characteristic of colloids that they were ar- rested by other colloids. H. Bechhold*^ called attention to the * [Graham recognized the slow diffusion of colloids. Tr.] 56 COLLOIDS IN BIOLOGY AND MEDICINE ability of true albumins to diffuse into gelatin jellies. This fact was demonstrated by means of the precipitin reaction. If goat serum is mixed with rabbit serum nothing noteworthy occurs. If the rabbit has previously been injected with goat serum and the serum of the previously treated rabbit (called ''goat-rabbit serum") is mixed with goat serum there occurs a precipitation of an albumi- nous substance, called ''precipitin." Bechhold mixed a 1 per cent gelatin solution containing 0.85 per cent NaCl with an equal vol- ume of goat-rabbit serum. The jelly was sohdified in the ice box and goat serum was layered over it. At the end of 24 hours a cloudy precipitate formed in the gelatin which in the course of 120 hours penetrated as far as 5 mm. The same phenomena occurred when the gelatin was mixed with goat serum and goat-rabbit serum was layered over it. Thus in both cases actual constituents of the serum had diffused into the gelatin. Similarly, Sv. Arrhenius* and Th. Madsen showed that not only diphtheria toxin and tetanolysin, but the highly colloidal diph- theria antitoxin and antitetanolysin could diffuse into 5 per cent gelatin jellies. Such a diffusion of colloids into a jelly naturally may be expected if the meshes are quite wide, i.e., if the jelly is quite dilute. Membranes. It will be appropriate to introduce the concept of "membrane" in the following way: if we make the colloid medium, the jelly, thicker and thicker, i.e., poorer in water, diffusion must be increasingly hin- dered. We very soon reach a point where no colloids are able to diffuse into it and we have reached a special case in our previous exposition, the membrane. We may describe membranes as irrever- sible gels, whose surface is very great in relation to their thickness. They play an important role in the organism, but we shall here discuss their general properties only, as their biological functions will be considered in Part III. An excellent general resume with a very complete bibliography has been published by H. Zangger ("Membranes and the Functions of Membrane"). On account of the great physical and chemical differences of the membranes of the organisms the employment of artificial membranes is preferable for the study of their chief properties. If a very dilute solution of potassium ferrocyanid is carefully layered over a concentrated solution of copper sulphate, there is formed at the layer of contact, by chemical interchange, a very thin brown film of copper ferrocyanid. Naturally this film is very deli- PHENOMENA OF MOTION 57 cate and is torn by the slightest movement. If we add gelatin to each solution and permit the two salts to diffuse towards each other in the jelly, there is formed at the layer of contact a very resistant membrane supported by the jelly. Expressed generally, if we per- mit two substances which form a precipitate together, to diffuse towards each other within a colloid medium which serves as a sup- port, a membrane is formed at the surface of contact which may have, depending upon the nature of the reacting substances, very different degrees of permeability. Since the time of Moritz Traube,* such membranes have been studied, especially by G. Tammann,* W. Pfeffer,*^ Adie,* P. Walden* and N. Pringsheim.* The chief interest, however, cen- tered in the osmotic phenomena of salt solutions which could be in- vestigated with the aid of such precipitation membranes, whereas the properties of the membranes themselves, with few exceptions, received but secondary attention. For investigations of osmosis the following substances are especially suitable: ferrocyanid of copper and ferro- cyanid of zinc; indeed all ferrocyanid-metal compounds are suitable since they are completely impermeable to many salts. They are briefly described as semipermeable membranes, because they are per- meable to water though impermeable to most crystalloids. If we permit a zinc ferrocyanid membrane to develop in a gelatin jelly and by the addition of potassium ferrocyanid exercise a very great osmotic pressure, the membrane will break in spite of the jelly sup- port, but it will not permit any potassium ferrocyanid solution to diffuse through. Besides this extreme case there are membranes of the most differ- ent permeabilities. Following up the work of the botanist N. Pringsheim,* H. Bechhold and F. Ziegler*^ exhaustively studied such membranes. They impregnated gelatin with silver nitrate or barium chlorid, and poured the molten solution into test tubes con- taining sodium chlorid or sodium sulphate. At times a layer of pure gelatin was interposed. At the surface of contact membranes of silver chlorid or barium sulphate were formed, which, however, were permeable for the salt solution on either side, because the mem- branes grew in the direction of the greater osmotic pressure, i.e., into the solution with the smaller osmotic pressure. If, for example, the silver nitrate solution was more concentrated it diffused through the membrane so that the latter grew into the sodium chlorid gelatin; but if the latter was more concentrated the reverse occurred. If both sides had the same osmotic pressure a very thin membrane formed which was sufficient however to arrest com- pletely the diffusion of both salts. Evidently the meshes of the net- 58 COLLOIDS IN BIOLOGY AND MEDICINE ^ work were filled with membrane-forming precipitate, for as soon as the membrane was melted, it again became permeable. In the same experiment, it was determined that diffusion was hindered only by visible precipitate membranes. There is no difficulty in forming by diffusion similar precipitate membranes from pure organic materials. We have already, on page 55, mentioned that a membrane may be formed by the diffusion of goat serum into goat-rabbit serum and we shall later refer to the fact that H. Bechhold*^ obtained membranes by the diffusion of metaphosphoric acid into gelatin containing albumin. There is no doubt that they may be obtained in other ways if desired. But it must by no means be assumed that a membrane is something rigid and unchangeable; on the contrary, it is constantly affected by the substances which flow through it and bathe it, making it more or less permeable, and in this way under certain conditions may evoke a self-regulation or a valve-like action. Hitherto there have been no investigations as to the manner in which the permeability of the precipitated membranes described is influenced by crystalloids diffusing through them. A priori it is to be assumed that such an influence exists just as in the case of re- versible jellies. That membranes may be more or less rapidly occluded by colloids is an observation which has been frequently made during the performance of ultrafiltration. A number of dried animal and vegetable membranes and parch- ment paper resemble precipitate membranes, in that they possess the same or but slightly superior swelling capacity. In dialysis they are used for the separation of colloids from crystalloids. Most of the membranes occurring in the organism are more or less swollen; on drying they lose this property to a great extent, as they are inelastic gels. Ultrafilters (see p. 95) formed by impregnating irreversible jellies are similar to natural membranes, since they must be kept in water to preserve their swollen condition. Membranes may be powerfully adsorbent, like reversible gels, and in this respect powerfully influence diffusion and flltration. Thus dyes, especially the basic ones, as well as certain groups of enzyms, e.g., arachnolysin, staphylolysin and rennin (H. Bechhold *4) are strongly adsorbed by many membranes. Such adsorbed substances may enter into chemical combination with the membrane (causing either shrinking or loss of swelling capacity) and thus diminish its permeability. This is the effect, e.g., of tannic acid, formaldehyd and Chromates. Alcohol, ether, acetone and sugar increase the permeability in cer- PHENOMENA OF MOTION 59 tain low concentrations; stronger solutions work to a certain extent in the opposite direction. The influence of electrolytes on the membranes of the organism and their permeability may depend on different causes. It may, for ex- ample, affect the swelling, and thus, the permeability. Alkalis in general increase the swelling; so also do acids in great dilution, but when concentrated they usually cause shrinking. Chemical changes due to chromic acid, etc., diminish the permeability. The influence of electrolytes is not however limited to this rather indirect action. We shall see on page 77 et seq., that differences of potential may develop at the interface between a solid phase and a fluid. Thus, for example, cellulose and wool are negatively charged with respect to pure water. In the case of the majority of animal membranes, most of which are amphoteric, the difference in potential develops only in faintly alkaline or faintly acid water. The pres- ence of salts likewise raises or lowers the difference in potential. The difference in the adsorbability of ions is accounted for also in this manner. Wherever a difference in potential exists, the diffusion rate of water is lowered. Salts, on the other hand, develop a differ- ence of potential in the course of diffusion; in their passage through a membrane, they raise or lower the existing difference in potential. •Vice versa, on this account the diffusion rate of salts is affected by the difference in potential existing in the membrane. A membrane may thus be the seat of an electromotive force (F. Haber,* Girard*), provided that it separates two salt solutions, or a salt solution from water, and that one solution be faintly acid or faintly alkaline. In the first instance the diffusion through the mem- brane will be much impeded, in the latter much accelerated. The results of R. Burian*^ also indicate differences in potential in the ultrafiltration of albumin-salt mixtures; at reduced pressures, he obtained as filtrate a salt solution isotonic with the liquid filtered. If he filtered under increased pressure, the filtrate contained a lower salt concentration than the original solution. F. G. Donnan* has made an unusually important and fundamental study of membrane equilibria, based upon the osmotic pressure and membrane potential of electrolytes containing a colloidal ion. I shall try to present Donnan's ideas without entering upon their mathematical basis. Let R represent an acid colloid, e.g., congo red, which forms a salt with Na, and let the line which separates the colloid electrolyte from water in our diagram be a membrane, impermeable to the colloid. (a) Membrane hydrolysis. Let us consider what occurs when the outer water is constantly renewed as described on page 92 et seq., which may be represented in the following diagram: Initial condition Terminal condition RH NaOH water RH water NaOH 60 COLLOIDS IN BIOLOGY AND MEDICINE i.e., there is formed by hydrolysis some colloid acid which cannot diffuse through, although the NaOH may do so. If the colloid is a strong acid, the process terminates rapidly providing the NaOH re- mains in the outer water, and an osmotic pressure may develop within the cell which is chiefly produced by the R and Na components of the colloid electrolytes as occurs, for instance, in the case of congo red. If the colloid electrolyte is a weak acid, then proportionally more NaOH diffuses outward, and when the equilibrium is established, it is chiefly by the hydrolytically split colloid acid and the NaOH. Ex- ample: soap solution. If the NaOH is constantly removed (by continually renewing the outer water, or by means of a bond which is not at all or slightly dissociated, e.g., carbonic acid) there will finally remain only colloid acid, in fact the weaker the colloid acid the more rapidly the process will terminate. What has been stated for an acid colloid applies of course to a basic one also. It follows from these premises that salts even of strong acids and bases may be completely broken up hydrolytically, provided one ion is a colloid which can be held back by a membrane. By membrane hydrolysis it is possible to separate from a neutral salt, either an alkali (intestinal or pancreatic juice) or an acid (hydrochloric acid, in the stomach, or acid urine). It requires no special exposition to show that the same process may be brought about by ultrafiltration. The reverse process may occur, however. If there is a colloid acid or base in a cell surrounded by a membrane, e.g., an amphoteric colloid albumin or fibrin, a minimal concentration of H or OH ions in the outer fluids suffice to form a salt with the colloid in the cell which by swelling develops a higher osmotic pressure. (h) We shall now consider what occurs when the colloid electrolyte within the membrane has an ion in common with the electrolyte outside, e.g., the Na salt of congo red (RNa) and common salt (NaCl). We then have the following formula: Initial condition Equilibrium R CI R CI Na Na Na Na (1) (1) CI (1) (2) Na ions cannot pass from space (1) to space (2) since by reason of its colloidal character the anion R cannot follow.^ However, CI and with it the same amount of Na will diffuse into (1). The amount of 1 There are always the same number of anions and cations in a solution. It is impossible to separate them by diffusion for then a free electric charge would be liberated. PHENOMENA OF MOTION 61 NaCl which diffuses depends on the concentration of the solutions in space (1) and space (2). If the concentration in space 2 (C2) is high in proportion to that in space 1 (Ci) much NaCl will pass from (1) to (2) and, if the conditions are reversed, only a little will do so. Mathematically represented (assuming complete electrolytic dis- sociation) we have the following equation, where Ci or C2 represent the molar ion concentration in the respective spaces and X the frac- tion of the molar ion concentration which diffuses from (2) to (1) : C2 — -A Ci -7- Co X Co X c If C2 is small in comparison with Ci we may express it : yr = tt' C2 Ci X 1 If Ci is very small the equation becomes jr = ?;' C2 2i The following table taken from Donnan's work illustrates the dis- tribution of NaCl. Original concentration of NaR in (1), Original concentration of NaCl in (2), Original relation of NaR to NaCl, C2 Per cent NaCl dif- fusing from (2) to (1), 100 X 0.01 0.1 1 1 1 1 1 1 0.1 0.01 0.01 0.1 1 10 100 49.7 47.6 33 8.3 1 Though we might assume d priori that the NaCl would distribute itself equally in both spaces in the presence of a membrane absolutely permeable for it, this table shows that the colloid electrolyte has a remarkable influence as soon as the concentration of NaCl falls. To a certain extent the colloid electrolytes drive the NaCl out of the cell. If Ci = 1 only about 11 per cent of a physiological salt solu- tion (C2 = 0.145) could penetrate the cell; or if it were already in the cell it was reduced to about 11 per cent. Apparently the mem- brane is permeable only from one side for the readily diffusible NaCl. (c) Finally we must consider the case when the colloid electrolyte in the membrane is opposed to an electrolyte without an ion in com- mon, as for instance: Initial condition Equilibrium Na K Na K R CI K Na (1) (2) CI R CI (1) (2) 62 COLLOIDS IN BIOLOGY AND MEDICINE Na will diffuse out, K and CI will diffuse in. Thus we obtain the following equation if CNa(i) expresses the molar concentration of Na in space (1). CNa(l) _ Ck(1) _ Cci(l) _ Ci + C2 _ CNa(2) Ck(2) Cc1(2) Ca If the concentration in the cell (Ci) is large compared with the outer solution (C2), then Ci -|- C2 _ Ci C2 C2 If Ci is small, then Ci + C2 _ . C2 Let us consider the equation representing a condition that frequently occurs physiologically. If Ci = 100 and C2 = 1 it follows that 99 per cent of the Na originally present in (1) will remain in (1) ; only 1 per cent will diffuse into (2). 99 p3r cent of the K originally pres- ent in (2) will diffuse into (1); only 1 per cent of the CI originally present in (2) will diffuse into (1). From this we may understand the hitherto inexplicable distribution of salt in cells, e.g., in red blood corpuscles. A colloidal anion attracts the foreign crystalloid cation and drives out the anions. Colloidal cations act in the reverse way. In conclusion Donnan derived formulas for the differences in electric potential which must exist after the equilibria described have been reached (membrane potentials). There are already many theories to explain the difference in poten- tial in organs and the electrical currents which occur in the body (muscle, nerve, electric-fish). These theories have the error that they require conditions which do not exist in the body, and in part that much greater differences in potential arise in the body than would be possible according to these theories. In this respect Donnan's theory differs much from its predecessors, f We shall not present his formula here but only a general explana- tion: If there are two equally concentrated solutions of NaCl sep- arated by *ä -membrane, and we insert a piece of platinum foil into each and connect them with a wire, no current will flow in the wire. If the solutions are of different concentrations, theoretically, electric energy wih be evident until the differences in concentration disappear as the result of diffusion. Such systems are called "concentration couples." A sy'stern consisting of colloid electrolyte, salt or water is a con- centration couple which develops a current in passing from "initial condition" to "equilibrium." This may be ascribed to the unequal PHENOMENA OF MOTION 63 concentration of ions on the two sides of the membrane. Let us consider the simplest illustration of the equilibrium between NaR and NaCl represented in our table on page 60 where the original con- centration is NaR : NaCl = 1:1 and equihbrium is established when 33 per cent NaCl passes from (2) to (1). The schematic representation would be: Equilibrium Na Na Na Na CI CI R CI (1) (2) In this instance all charges are mutually satisfied excepting those of CI and R. The slowly diffusing anion R is opposed to the rapidly diffusing anion CI so that a difference of potential must arise at the boundary surface. In the cases hitherto described, the membrane itself is the seat of the difference in potential. The conditions are quite different if the membrane acts only as a bounding surface, that is, if it is not equally permeable for all ions. In this case, the uni- versally present "contact potential," existing in two contiguous salt solutions, is modified by the aforementioned property of the membrane. Finally, we must recall another kind of membrane which does not fall into any of the previous categories. According to W. Nernst, a film of water upon ether forms a semipermeable membrane for benzol. The experiment is carried out in this way: A pig's bladder is soaked in water; the bladder plays no part, other than to hold the water which forms a partition between ether and ether containing benzol. Here the semipermeability of the membrane depends en- tirely on selective solubility. Benzol is insoluble in water; ether on the contrary has a limited solubility, and as a result ether diffuses through the water to the benzol. Subsequently many such com- binations were devised. They are very extensively distributed in the organism. It is unnecessary to think of complete semipermeability in every case; scattered deposits (fat, lecithin, etc.) nraj^^suffice to bring about a partial permeability. The membranes of Wistinghausen depend on this principle of selective permeability. He impregnated with gallic acid salts, animal membranes which then became permeable for fat; by merely washing away the salts the permeability is abolished. Attention should be called to a remarkable observation of Zott (cited by H, Zangger) which belongs in this chapter. He discovered that a membrane through which sugar has diffused, permitted the passage of gum arable after it had been moistened with alcohol. CHAPTER V. CONSISTENCY OF COLLOIDS. Internal Friction. The various colloids show all possible transitions from fluids to solid substances. A fluid may take on any shape, and the work necessary to change its form, i.e., to overcome its internal friction, is very slight. Solid substances, according to Wilhelm Ostwald, pos- sess a form-energy also called elasticity; the energy necessary to change their form, i.e., to overcome their internal friction, is very great. If we picture to ourselves a number of colloids and gels we pass from a true fluid, water for instance, through the albumoses and albumin solutions to the semifluid gels {e.g., 1 per cent gelatin), jellies and finally to the firm substances {e.g., horn). High internal friction, viscosity, is a typical property of hydrophile colloids. Since colloids are diphasic systems, the internal friction will depend, above all, upon the size of the free surface of the colloid, i.e., upon the concentration. Changes in temperature are of great importance. The absolute as well as the relative influence of concen- tration is, indeed, characteristic for colloids. Even traces of colloids (agar, gelatin) may increase the viscosity of water to an extraordi- nary extent. We may obtain all degrees of internal friction with agar and in fact a 5 per cent solution of agar is a solid body at room temperature. Usually the viscosity increases with decrease in temperature, inas- much as substances then approach the solid condition. Gelatiniza- tion is analogous to the sohdification of a molten fluid, where internal friction rapidly rises within a small temperature range. J. Friedländer,* D. Holde* and V. Rothmund* proved that artificial emulsions (gum water, castor oil, so-called solid fats) exhibit a variation in their viscosity curves according to temperature and concentration, similar to that shown by many natural hydrophile colloids. T. B. Robertson* found that emulsions of oil in water became increasingly more viscous the higher the concentration of the oil, until a critical point was reached when the viscosity decreased; the water then became the dispersed phase. Internal friction is indeed a very complicated phenomenon. It depends according to W. B. Hardy upon (1) the internal friction of 64 CONSISTENCY OF COLLOIDS 65 the different phases, (2) the surface friction of the internal surfaces, (3) the surface tension of the internal surfaces and (4) the strength of the electrical charge. To what extent the individual factors influ- ence the internal friction is as yet unknown. We have, however, received valuable guidance from the study of the effect of electrolytes. Especially remarkable is the parallelism between swelling and internal friction. It depends, apparently, on the fact that both phenomena are characterized by an increase, i.e., multiplication, of the free surfaces. Thus, for instance, we see that acids and alkalies which favor the swelling of gelatin also increase the internal friction of albumin, because there probably occurs an in- crease in the free surfaces of the albumin ions (see p. 153 et seq.). Swelling and Shrinking. If a crystalloid (common salt, sugar) is thrown into water, it sub- divides in it until finally it is completely dissolved; the particles of salt or sugar lose their cohesion. A colloid (glue, wood) in contact with water increases its volume, it swells; its particles retain their cohesion. This property, however, is possessed only by hydrophile colloids. The imbibition of water, that is, swelling, may either go on in- definitely in the case of colloids, so that finally the particles are torn asunder and a solution or sol is formed as in the case of albumin; or the imbibition may reach its limit very rapidly, as in the case of wood. Between these there are all sorts of transitions, e.g., glue. Only in the case of gels is it usual to refer to swelling. In the organism, gels having very slight ability to swell serve as covering and framework {e.g., hide, collagen, shells and wood). They are intended to retain the outward form. The same is true of the supporting tissues of the individual organs and even of the cells, vessel walls, the mem- branes of the intestinal canal, connective tissues, the vascular bundles of plants, cell membranes, etc. On the other hand, the cell content possesses the ability to swell to a high degree. Every organ has a certain definite normal fluid content. A healthy plant has a definite turgescence and the protoplasm of a healthy animal a given degree of swelling; every abnormal change in this signifies illness or even death. Without doubt swelling plays a very important role in the case of many phenomena which have hitherto been attributed to osmotic pressure. Indeed, the osmotic pressure is only manifested completely in the presence of a semipermeable membrane, whereas the ability to swell does not require the presence of a membrane. Swelling may under some circumstances counter- balance the osmotic pressure or even overcome it and concentrate 66 COLLOIDS IN BIOLOGY AND MEDICINE solutions. An excellent example of the latter is described by C. Ludwig. He hung a well-dried animal bladder in concentrated salt solution. The bladder swelled, taking up a dilute salt solution and common salt crystallized out in the remainder. Amphibia, e.g., frogs, may lose one-fourth of their body weight upon drying as has been shown by E. Overton.*^ Although they contain about 80 per cent water, the osmotic pressure of the blood almost doubles. The explanation of this is that only a portion is water of solution, the remainder is water of swelling. Upon diying the water of swell- ing is more strongly retained than the water of solution. Swelling exhibits manifestations of energy to no less a degree than osmotic pressure. I shall give several examples taken from Wolf- gang Ostwald's "Grundriss." According to the investigations of the plant physiologist, Hales, swelling peas were able to lift the cover of an iron pot weighted with 83.5 kilograms. H. Rodewald found that it requires 2523 atmospheres pressure to overcome the swelling pressure of starch. J. Reinke* determined the swelling pressure of laminaria, a sea weed. Some of his data quoted from H. Freundlich give us an idea of the enormous pressures, changes in volume and amounts of water taken up when swelling occurs and of the pressures required for dehydration. Ten layers of dried laminaria scales each 0.1 mm. thick and 50 mm.^ were placed in the apparatus. Pressures in atmos- pheres. Elevation in mm. due to swelling. Pe-centage of water by volume in air-dried substance. 41.2 21.2 7.2 1 0.16 0.35 0.97 3.30 16 35 97 330 We obtain a fair idea as to the general course of swelling by ob- serving a sheet of gelatin. Dry gelatin takes up one-third of its weight of water from a moisture-saturated atmosphere at room temperature, in order to reach a condition of equilibrium. If this sheet is then placed in a dish of cold running water it absorbs from it 10 times its dry weight of water in order again to reach a condition of equilibrium. In dry air the water evaporates and shrinking occurs. The experi- ment may be repeated as often as desired with the same result. On this account substances of this group are termed elastic gels. Coagulated albumin, e.g., boiled fibrin, behaves differently. If it is air-dried a horny residue remains which, though it takes up some water, or swells when it is placed in water, never again CONSISTENCY OF COLLOIDS 67 approaches its original gelatinous state; such gels are called inelastic gels. One of the most important operations of microscopical technic is the hardening of elastic gels. (See Chapter XXIII.) Their ability to swell in water is destroyed by the chemical action of formalin, chromic acid, mercuric chlorid, etc. We must consider that in the organism, the inelastic gels, i.e., connective tissue and also the cell pellicle, etc., arise from elastic gels by chemical changes with con- sequent loss of water or drying, as may be observed at the surface of any wound. In this cormection, we may revert to the formation of surface films (see p. 33), whose formation is certainly more than merely analogous to that of organized membranes, skin, etc. Classical studies on the swelling and shrinking of slightly elastic gels were made by J. M. van Bemmelen in the case of silicic acid gel and amplified by 0. Bütschli, So many difficulties are unfor- tunately offered to the application by analogy of these properties to organized inelastic gels, that we must confine our attention to the most important ones. The evaporation of water from a silicic acid gel proceeds at first as it would from a solution. When the gel reaches a certain consistency a turbidity appears, that is, hollow spaces of about o fifj, form between the supporting walls of the gel which become filled with air. Upon losing still more water, the tur- bidity disappears and the gel becomes glassy. In this latter respect the inelastic gel of silicic acid differs very materially from the elastic gel of gelatin, which does not become turbid. This is likewise the case in the reabsorption of water. Though gelatin shows a similar curve both on swelling and shrinking, silicic acid gel and indeed we may say all inelastic gels show entirely different curves. That is, the swelhng of elastic gels is practically completely reversible, whereas with inelastic gels this is not the case. The changes a gel undergoes on freezing and thawing are very similar to those of shrinking and swelling. The crystallization of ice from a gel containing water indicates a withdrawal of water, whereas upon thawing, water becomes again available for swelling (H. W. Fischer, 0. Bobertag and C. Feist*). There are consequently substances which after freezing and thawing revert almost completely to their original state, e.g., soluble starches, fish glue, whereas others, e.g., silicic acid hydrosol and albumin, undergo changes which are more or less irreversible. The influence of electrolytes on the swelling of gelatin, agar, pig's bladder, cartilage and fibrin, is very considerable. It has been in- vestigated especially by F. Hofmeister,* Wo. Pauli, *i K. Spiro,* Wo. OsTWALD,*^ and Martin H. Fischer.* It may, in general, be stated that acids and alkalis increase the swelling capacity to an 68 COLLOIDS IN BIOLOGY AND MEDICINE extraordinary degree. This, however, does not depend only upon the electrolytic dissociation of various acids and the concentration of H or OH ions. In the case of strong acids it reaches a maximum at a certain concentration and then decreases. Thus Martin H. Fischer found that fibrin, which swelled to 8 mm. in water, in 0.02 normal HCl reached the maximum swelling of 48 mm. ; whereas in 0.1 normal HCl the swelling reached only 21 mm. In the case of H2SO4 the maximum swelling was only 11 mm. in 0.024 normal acid. Purified glutin (according to the experiments of R. Chiari in Pauli 's Fig. 7. The swelling of fibrin in solutions of various sodium salts (^V molecular). (From M. H. Fischer.) laboratory) is so sensitive to acids that it swells less in distilled water than in Vienna Hochquellwasser, because of the CO2 contained in the latter. Furthermore, distilled water may even be distin- guished from conductivity water by swelling experiments with glutin. The swelling in alkalis is still greater; in 0.02 normal NaOH it reached 77 mm. M. H. Fischer believes that the swelling in acids is dependent upon the concentration of the H ions minus the effect of the anions of the acid under consideration. In this case, there probably exists an antagonism between cation and anion, such as may be demonstrated in the case of neutral salts. A similar rule probably obtains for alkalis. CONSISTENCY OF COLLOIDS 69 When such heterogeneous substances as gelatin, fibrin, etc., behave similarly under the influence of electrolytes, we may assume that the same cause determines the behavior, and this cause must be sought in the chemical nature of these substances. Apparently these sub- stances are amphoteric, i.e., at the same time weak acids and weak bases, forming under the influence of acids and bases, more or less ionized salts. Ionization causes an hydration, i.e., an imbibition of water which is evidenced in the case of these gels by their capacity to swell, and in the case of dissolved albumins, by an increase of the internal friction. We should therefore not expect to see these phe- nomena in the case of gels of entirely different chemical properties, silicic acid gel, for example. Neutral salts, to a certain extent, favor the imbibition of water and indeed the swelling is greater in such dilute salt solutions than it is in pure water. At a certain concentration (for NaCl, 13.8 per cent) the amount of fluid taken up reaches a maximum and then falls again. The anions are primarily active in favoring swelling, whereas the cations have a lesser influence and, in favoring swelling, CNS > I > Br > NO3 > CIO3 > CI; whereas in favoring shrinking, SO4 > tartrate > citrate > acetate. Although a dry jelly removes more water from a salt solution than it does salt so that the concentration of the solution is increased, a swollen jelly takes up more salt than it does water, thus diminishing the concentration of the solution. The swelling in acid or alkaline solution is always much decreased hy the presence of neutral salts, and anions are much more active than cations. In producing this decrease citrate > tartrate > phosphate (?) > SO4 > acetate > I > CNS >N03>Br>Cl; Fe > Cu > Sr > Ba > Ca > Mg > NH4 > Na > K. Though, e.g., 0.78 gm. gelatin in 100 cc. of 0.05 nHCl swelled up to 14.61 gm., the swelling reached only 2.84 gm. in the presence of -^ potassmm citrate and about 7 gm. in the presence of -^ KCl. Antagonistic Effects. In addition to the antagonistic action of neu- tral salts on the swelling due to H and OH ions, there is also an antagonism to monovalent cations hy 'polyvalent cations. 70 COLLOIDS IN BIOLOGY AND MEDICINE _ Thus it has been determined by the experiments of Martin H. Fischer on fibrin and of Wo. Ostwald on gelatin that swelHng is much more strongly depressed by polyvalent cations than by mono- valent ones (Mg < Ca < Ba < Sr < Cu < Fe) and it seems probable that polyvalent cations counteract the action of monovalent ones in favoring swelling. So far as I know there is as yet no colloid chemi- cal confirmation of this assumption in the case of swollen colloids. There are, however, a number of biological experiments concerning the inhibition of the poisonous action of neutral salts by polyvalent cations (see p. 378), which in all probability are referable to the in- hibition of harmful swelling. According to these biological experi- ments the antitoxic effect of cations increases with their valence and stands in relation to the ionization pressure or the electrolytic solu- tion tension.^ Our present knowledge indicates that the swelling and the shrink- ing of hydrophile gels absolutely parallels the formation of ions and neutral particles in the case of albumin. This has been more ex- haustively discussed on page 153 et seq. The same factors which favor the ionization of albumin, namely, acids and alkalies, also favor swelling. In this case as in the other the presence of neutral salts depresses the action of acids and alkalies, polyvalent cations or anions acting more powerfully than monovalent ones. We rec- ognize in both ionization and swelling a tendency towards an in- crease of the free surface, which is associated with the taking up of water. This may go so far that the molecules are split; hydrolysis occurs and cleavage products are formed. Accordingly, chemical re- actions, especially hydrolytic cleavages, occur much more rapidly in swollen than in shrunken colloids. For instance (according to E. Knoevenagel*), the hydrolysis of swollen acetyl-cellulose by potassium hydrate requires only a few minutes; but the same process requires days in the case of the shrunken material. Non-electrolytes have only a slight influence on swelling. Of the few cases known to us we may mention that urea favors the swelling of gelatin even in acid solution, but it has no effect on fibrin. Alcohol and sugar favor the swelling of gelatin in a certain concentration be- tween 1 and 2 per cent. These data were almost exclusively obtained with gelatin and fibrin. Both gels behave qualitatively alike (excepting with urea) though there are quantitative differences. Fibrin swells much more ^ To avoid any misunderstanding it should be stated that substances which are themselves strongly toxic, e.g., barium, zinc or lead salts, may act in suit- able small doses as antidotes (probably by counteracting swelling) to harmful quantities of neutral solutions {e.g., pure NaCl solutions). CONSISTENCY OF COLLOIDS 71 than gelatin. Gelatin may absorb about 25 times its weight in water; fibrin 40 times. The order in which neutral salts act on gelatin is different from that in which they act upon fibrin. It may be assumed that the different gels of the organism vary quantitatively in their behavior under the influence of the same elec- trolytes and it is obvious that the salt absorption of different gels varies as well as their water absorption. An investigation of the water and salt absorption of different kinds of gels in the presence of mixtures of electrolytes is much to be desired. Our knowledge of tissues and secretions forces us to the conclusion that the different tissues possess a very different specific ability to absorb certain sub- stances or ions. Only thus can we understand why the blood cor- puscles withdraw more potassium salts from the lymph, the cartilages more sodium salts and the bone-building tissues more calcium salts. Only thus can we obtain a conception of the specific crystalloid con- tent of various secretions and of selective resorption. The Crystallization of Colloids. Though P. P. VON Weimarn describes the crystalline state as the "sole ultimate condition of matter" which is characteristic^ for all substances (even gases), we shall not attempt here a critical study of this theory nor determine the limits of the crystallization of solids. We shall consider only how crystals occur in colloidal substances and more particularly limit ourselves to the biocolloids. We know only a limited number of crystallizable biocolloids; the most im- portant are egg albumin, horse serum albumin, certain plant albumins (aleuron crystals from Para nuts, cotton, hemp and sunflower seeds), oxyhemoglobins, hemoglobin and methemoglobin. Egg albumin has been obtained in the shape of needles, the albumins of vegetable seeds partly in octahedra and partly in tablet-shaped hexagonal prisms. Oxyhemoglobins crystallize in various ways depending upon the animal species from which they are derived. For example, horse oxyhemoglobin forms rhombic, and squirrel oxyhemoglobin forms hexagonal prisms. These substances may be recrystallized and under the same conditions give the identical crystal form. It may be re- marked in passing, that many crystals giving the albumin reaction have frequently been observed in organs, but they have been in- sufficiently studied. [Crystalline form is markedly influenced by the presence of protective colloids in the crystallizing solution. See J. Alexander, Kolloid Zeitschrift, iv, p. 86. Tr.] Crystalline products have been obtained from starches, e.g., sphero-crystals from inulin. ^ Bibliography given in Wo. Ostwald's Grundiss der Kolloidchemie (Dresden, 1911). 72 COLLOIDS IN BIOLOGY AND MEDICINE The crystallization of alkaline salts of the higher fatty and arylic acids is well known. The crystals of colloids are distinguishable from those of crystal- loids in many respects. That their solution is preceded by a swelling is not surprising in view of the hydrophile colloidal character of the substances under consideration. On the other hand, it is remarkable that other constituents may always be demonstrated as inclusions. The crystallized globulins from vegetable seeds always contain com- mon salt. K. A. H. MoRNER* showed that only the sulphates of egg and serum albumin were crystallizable. As E. Abderhalden*^ has shown, oxyhemoglobin crystals do not contain their proper pro- portion of albumin, and although numerous researches on crystallized egg albumin have been undertaken, in different instances the amount of contained carbohydrate varied. In spite of these facts, we are of the opinion that colloids can actually crystallize, and that their crystal form is not controlled by the crystalloid impurities. We know that crystalloids frequently include mother liquor, that they may form mixed crystals and that it is often impossible to remove impurities by ordinary re- crystallization. In view of the persistent salt content of crystallized albumins it is probable that only their salt-like compounds possess a definite crystalline shape. Especially favorable to this view is the fact reported by Dabrowski, that crystallized egg albumin, when placed in a 3.6 per cent solution of ammonium sulphate, exhibits a more rapid diffusion than salt free egg albumin, and has about one-sixth of the atomic volume of the latter. The crystallized egg albumin, therefore, is formed of smaller particles. The Life Curve of Colloids. Though in the absence of chemical changes, crystalloids retain their physical properties, in the case of colloids after a lapse of time changes occur which are conunonly called aging. For instance, silicic acid which has been freshly prepared from water-glass solution and HCl is at first dialyzable but loses this property after a few days. Most of the "aging phenomena" of sols are characterized by the fact that the particles of a highly dispersed solution gather together to form larger particles, that their sensitiveness to flocculation is increased or that they spontaneously coagulate. In the case of gels, their elasticity suffers changes and they become optically inhomogeneous or turbid. Bearing in mind that colloids are metastable systems, it is obvious that in the course of time they must change, since they tend to be- come stable systems. In the examination of a colloid, the properties CONSISJ'ENCY OF COLLOIDS 73 found, strictly speaking, are applicable to its momentary condition; previously and subsequently it has different properties. Every point of its life curve has a previous history and the final portions of this constantly flattening curve are the aging phenomena. In con- tradistinction to crystalloids every colloid is a particular individual. If solutions of hydrophobe colloids, e.g., arsenic sulphid, gold solu- tion, etc. (without protective colloid), are permitted to stand for some time, they flocculate after a short time or else after a lapse . of years. It may be that traces of electrolytes are responsible for the flocculation. In other cases, electrolytes certainly play no part; as I shall show by a number of examples, there is an evident tendency for the unstable colloids to pass over into less dispersed and stable systems (see L. Wöhler's* observation on the aging of col- loidal molybdic and tungstic acid). Several years ago H. Bechhold and J. Ziegleri sought to prepare for therapeutic purposes, with the aid of new and especially suitable protective colloids, solutions of such organic substances as are insoluble in water (iodoform, iodo- chloroxychinolin, camphor, etc.). They succeeded in thus preparing the substances, which, however, kept only a few weeks, when they would separate out in crystals. Obviously these substances are not sufficiently insoluble and they exhibited the adsorption phenomenon de- scribed on page 18. P. P. von Weimarn made analogous observations on the sol of barium sulphate in which crystals appeared at the end of six months. The inequality of the particles, or more correctly the ''specific surface," obviously militates against the stability of such colloid solutions. In the majority of cases, it soon leads to the "death" of the colloidal system. Furthermore, we must emphasize that the changes in the col- loidal system need not always consist in a diminution of the disper- sion. Occasionally we find that the particles become smaller with the lapse of time, but this has hitherto been observed only in the case of hydrophile colloids (glycogen, benzopurpurin, hemoglobin, lecithin, etc.) (W. Biltz and L. Gatin-Gruszewska,* Lemanissier,* E. RÄHLMANN,*! R. Zsi GMÜND Y*2). Under some circumstances even electrolytes may act disruptively. Thus B. G. Moore and H. E. Roaf* observed that minutest traces of electrolytes are absolutely necessary for the stability of an albumin solution, as was frequently pointed out by E. Jordis for hydrophile sols. W. Biltz and H, von Vegesack* observed, however, that in the case of dye solutions, merely with the lapse of time, marked in- crease in viscosity occurs. It may be pointed out in connection with the aging of jellies, that ^ As yet unpublished. 74 COLLOIDS IN BIOLOGY AND MEDICINE freshly poured gelatin cylinders reach a practically constant modulus of elasticity at the end of three to four hours. This accords with the fact observed by F. Stoffel, that crystalloids diffuse more rapidly in quickly chilled than in slowly chilled gelatin, and that this differ- ence disappears after several days (see p. 54). At the outset we spoke of the "life curve of colloids," of "aging phenomena," "death," "individual properties," etc., and it might appear that these are only similes borrowed from the organized world. In my opinion the relationship is closer, and I believe that we may obtain a more profound understanding of the phenomena of Life (so unintelligible to us) by a study of such phenomena in colloids. Aging has hitherto been considered, for the most part, a purely biological phenomenon. In my opinion, we may attack the problems with the methods of exact science, if we could but separate two groups: the organs (cell groups), which constantly renew themselves, from those which are lasting. We would, a priori, expect changes in the latter similar to those observed as aging phenomena in col- loids. We saw that a rapidly chilled gelatin was at first easily pene- trable for crystalloids, but that with time its resistance increased. We may, therefore, assume that in young organs (fresh membranes) the exchange of matter by diffusion proceeds more rapidly. The decrease in elasticity, one of the most characteristic phenomena of aging, may be measurably followed in aging gelatin. In fact it has been shown that for the vital staining of nerves with methylene blue, young animals are more suitable than old ones. With aging there occurs shrinking, which begins already in intrauterine life. In the third month of human fetal life the water content is 94 per cent, at birth it is 69 to 66 per cent, in adult life 58 per cent. We may say in general that with aging there is a decrease in the swelling capacity of the organ colloids. This holds both for animal organisms, which lose water as they grow older, and for plants (dry leaves — lignifi- cation). Tyndall Phenomenon. (From Wo. Ostwald.) J (a) (b) Suspensions of Lamp-black, (a), uncoagulated; (b), coagulated. (From E. E. Free.) PLATE I. CHAPTER VI. OPTICAL AND ELECTRICAL PROPERTIES OF COLLOIDS. Optical Properties. Colloidal solutions, e.g., albumin, always show a slight turbidity. If a strong ray of light is passed through such a solution, its path may be seen as a bright band. (See Plate I.) The "Tyndall phe- nomenon," as it is known, is much more distinct if a ray of light is passed through smoke or through a turbid suspension, in which case, the reflected hght is polarized. This phenomenon manifests itself if a sunbeam passes into a dark room. The illuminated dust par- ticles (motes) appear bright against the dark background. Michael Faraday observed the phenomenon in the case of gold hydrosols and he was led to the opinion that such solutions, which we nowadays call colloids, were nothing but extremely finely divided suspensions (dispersed systems). It was a great service to science, when H. Siedentopf and R. Zsigmondy recognized the importance of this phenomenon for the investigation of colloids, and constructed an instrument adapted to this purpose by passing the reflected light into a microscope. In this way they obtained bright pictures of the suspended particles on a dark ground. Since neither of these in- vestigators paid any attention to the representation of the shape of the particles, and devoted their attention only to the reflection of a point of light, it was possible, by utiHzing the strongest sources of light (sun, arc lamp), to perceive particles lying below the limit of microscopic visibility. They, therefore, called the apparatus the ultramicroscope . The great number of fundamental observations with the ultra- microscope which we owe to R. Zsigmondy *2 and his followers have been repeatedly mentioned. R. Zsigmondy called the particles which he could definitely distinguish against the dark field, but which were far below the limit of microscopic visibility, submicrons (from 6 to 250 fxix). If only a faint cone of light could be seen, it is to be assumed in many cases, that the smallness of the individ- ual particles precludes the recognition of each one. Such particles (under 6 mm) be called amicrons. Most inorganic hydrosols, especially metals, form characteristi- cally colored solutions, e.g., silver hydrosols are brown, platinum 75 76 COLLOIDS IN BIOLOGY AND MEDICINE hydrosols are greenish brown, gold hydrosols are red but become blue and finally brown when electrolytes are added. In the ultramicro- scope the individual particles are not of a uniform color. For in- stance, coUargol has blue, red, violet and green particles, the particles of a red gold solution are chiefly green, those of blue solutions range from yellow to reddish brown. Theoretically, submicrons of the same size should have the same color so that the variety of color in the ultramicroscope indicates variation in the size of the particles. As a matter of fact, as we have said, the smaller submicrons of finely divided red gold hydrosol are almost all green though there are very small brown submicrons. There is no entirely acceptable explanation for the color variation of submicrons of identical size. The number of particles visible in the ultramicroscope is, in the case of hydrophile colloids, usually far less numerous than might have been expected from their other properties. This is the result of their inferior reflecting power. If a piece of swollen gelatin is immersed in water it becomes invisible, because no light is reflected to the eye. On this account the ultramicroscope is not suitable for determining the number or size of the particles of hydrophile colloids. We may here recall an observation of G. Quincke*^ which is perhaps destined to be of great importance for many biological questions but which deserves attention, even from a purely physical standpoint. G. Quincke observed that in the induced clarification of mastic, gamboge, kaolin and India ink suspensions, the flocks usually separated on the dark side; in spontaneous clarification of kaolin turbidity, however, they settled on the light side. A turbid solution of tannate of glue settled out, mostly on the side towards the light. He described this phenomenon as positive and negative photodromy. This fact is suggestive of many of the phenomena which H. SiEDENTOPF*^ observed in his light reactions in the ultra- microscope. Jellies, especially those of higher concentration, on deformation by compression or traction, show double refraction (Kundt). Nega- tive refraction was observed in the case of gum arable, collodion and gelatin, positive in the case of tragacanth and cherry gum. The same kinds of jellies when dried showed respectively the same kinds of refraction. If gelatin poor in water is brought in contact with gelatin rich in water, so that there is a mutual interchange of water, both become doubly refractive. (M. W. Beijerinck.) On re- peated swelling and shrinking of jellies, the positive double refraction passes through an isotropic condition into a negative. (Quincke.) OPTICAL AND ELECTRICAL PROPERTIES OF COLLOIDS 77 Chlorids and nitrates diminish the double refraction; sulphates are without effect. Phenols change the direction of the refraction. The phenomenon is important for the understanding of the double refraction of organized structures (plant fibers, muscle, horn, etc.). Electrical Properties. If two electrodes are placed in a solution of a hydrosol as free as possible of electrolytes, and a current is allowed to pass through, we immediately notice the movement of the colloid to one of the elec- trodes. Various suspensions (suspensions of clay, rosin, etc.), as well as most hydrophobe hydrosols migrate to the anode, whereas the colloidal metal hydroxids (iron or aluminium oxid hydrosol, etc.) move to the cathode. Hydrophile colloids (albumin, etc.) exhibit, if almost free from electrolytes, no definite recognizable directive tendency. The zone of H ion concentration in which there is no migration has been named by L. Michaelis the isoelectric zone. The addition of acids causes migration of these colloids to the cathode, alkalies to the anode; they then behave as if they were salts of the acids or alkalies involved, and this may actually corre- spond with the facts (see p. 149 et seq.). This movement of suspensions and hydrosols against the water, under the influence of the electric current, is called cataphoresis. The colloid particles behave like ions and their speed of migration is similar in rate. Zsigmondy calculated from the speed of migration (0.002 mm.) and the diameter (50 mm) of the particles of a colloidal silver solution, that the particles were charged with 297 X 10"^" electrostatic units which is the equivalent of 62 elemental units. Such a particle is in a certain sense an ion of 62 valencies. If a protective colloid (albumin, gelatin, etc.) is added to a sus- pension, the latter acts as if its entire mass was composed of the protective colloid. The commercial inorganic colloids (collargol, lysargin, etc.) do not behave in an electric current as does pure colloidal silver, but as albumins or albumoses. Their direction may be changed at will by the addition of acids or alkalis. The process may be reversed, that is, the water may be moved under the influence of electrical difference in potential, provided the suspension is held fixed. The experimental procedure is as follows: Instead of a clay suspension we choose a porous clay wall D (Fig. 8), permeable for water, by means of which a U-shaped tube is divided into two parts. If the tube is filled with water and into each branch an electrode is introduced, the water moves under the influence of the electric current and, in fact, it will rise on the cathode side until it exerts a certain pressure against the anode 78 COLLOIDS IN BIOLOGY AND MEDICINE side. This process is called electro-endosmosis. In principle the two phenomena are the same. For the exact study of electrical migra- tion, which was first investigated by G. Wiedemann and G. Quincke, electro-endosmosis proved to be experimentally more easily avail- able. A. CoEHN* has shown that it is a general and quantitative law, V ^S -^-^. jy Fig. 8. Apparatus for electro-endosmosis. (H. Freundlich.) that substances possessing higher dielectric constants are positively charged when brought into contact with substances of lower dielectric constants. We have already seen that numerous substances are negatively charged in respect to water while others are positively charged and that under the influence of the electric current there occurs a move- ment of the suspended substance or of the water. This migration may be influenced by the addition of electrolytes. J. Perrin used diaphragms of porous carborundum and of naphthalin and measured the amount of water which passed through the porous wall by en- dosmosis, under the influence of an electric current. In the following table + indicates passage to the positive, — to the negative pole. Diaphragm. Solution. Amount of fluid transferred in cmm. per minute. Carborundum Carborundum Carborundum Carborundum Carborundima Carborundum Naphthalin ^ 0.02 mol. HCl 0.008 mol. HCl 0.002 mol. HCl dist. water 0.0002 mol. KOH 0.002 mol. KOH 0.02 mol. HCl 0.001 mol. HCl 0.0002 mol. HCl 0.0002 mol. KOH 0.001 mol. KOH + 10 0 - 15 - 50 - 60 -105 + 38 Naphthalin Naphthalin + 28 + 3 Naphthalin. . - 29 Naphthalin . - 60 I I tried to cause naphthalin suspensions to migrate in strong currents (400 volts) but could not observe any cataphoresis either in neutral, in acid or in alkaline solution. Nor could I obtain a migration of suspensions of naphthol and naphthylamin in neutral, very faintly acid or very faintly alkaline solution (hitherto unpublished). OPTICAL AND ELECTRICAL PROPERTIES OF COLLOIDS 79 It follows from this that the negative charge of negative diaphragms (as is evident from the table) increased in alkaline solution. With a decrease in the OH ion concentration (with naphthalinj or the in- crease in the H ion concentration (with carborundum) a point was reached in which there was no difference in potential between w^ater and diaphragm. With further increase in the H ion concentration the diaphragm took on a stronger positive charge. If a positive diaphragm is chosen, as for instance chromium chlorid, the conditions are reversed. J. Perrin studied the influence of salts in the presence of acids and bases. It was sho^^^l that they cause a loss of charge with moder- ate concentrations and, in fact, that the strength of their action de- pended, in the case of positively charged diaphragms, upon the valence of the anions, while in the case of negatively charged dia- phragms it depended upon the valence of the cations. With higher concentration of polj^^alent anions or cations a loss of charge may occur. Upon calculating the concentration of salt which just halves the amount of fluid (r in mm./minutes) transferred, the follo'^ing figures were obtained: Diaphragm. Charge. Salt. V 1 = 1 (NaBrorKBr). Carborundum XaBr 50 1 Carborundum — Ba(X03)2 2 25 Carborundum — La(X03)3 0.1 500 Chromium chlorid + KBr 60 1 Chromium chlorid + ]MgS04 1 60 Chromium chlorid + K3Fe(CX)6 0.1 600 The last column - shows us that the discharging effect increases V with the valence of the anions as 1 : 25 : 500 and in the case of cations as 1 : 60 : 600. In the section on " flocculation, " we shall see a vers" remarkable application of this phenomenon. Of the many theories proposed to explain these circumstances, that of H. Freundlich and L. ]\Iichaelis, according to which the various ions are adsorbed to different degrees, seems to us most prob- ably correct; in point of fact, the H and OH ion have especially high adsorption coefficients. Since a separation of anion and cation usually cannot occur, there arises a difference in potential at the surface between the dispersed phase and the water. An indifferent substance in a weakly acid solution will adsorb H ions and become 80 COLLOIDS IN BIOLOGY AND MEDICINE positively charged in respect to the fluid; in an alkaline fluid it will become negatively charged. If the dispersed phase itself has basic or acid properties it will behave in pure water like a cation or an anion respectively. If to an acid suspension, e.g., clay, an alkali is added, K ions are adsorbed, OH ions are concentrated and held at the outer film and the negative charge is thereby increased. The reverse occurs upon adding acids; the charge is released and may even take the opposite sign. According to this view, colloids wan- dering to the anode are discharged only by cations; those travelling to the cathode only by anions. We have seen that polyvalent ions have a considerably greater discharging power, which increases with their valence. This experimental fact accords, as do all the others, with the theory propounded. In the case of hydrophile colloids, it suffices for us to assume that we are dealing with very large amphoteric molecules which become cations in acid solution and anions in alkaline solution. Salting Out. If large quantities of a neutral salt, for instance ammonium sulphate, are added to a solution of a hydrophile colloid (albumin, globulin, casein, albumose, silver protected by a protective colloid, etc.), or certain inorganic hydrosols, e.g., sulphur, the colloid is thrown out, but it redissolves upon dilution with water. This is the process of salting out, as practised technically. If, for instance, enough com- mon salt is added to an aqueous solution of phenol, the phenol separates out. We might regard this as a withdrawal of water by means of the electrolyte, since we know from the observations of recent years that ions form hydrates in aqueous solution, i.e., every ion attracts a greater or smaller number of molecules of water. I have, however, been unable to discover any relation between the salting out process, from the figures for the hydration of a number of different ions as calculated by E. H. Riesenfeld and B. Reinhold*. The more closely a hydrophile colloid approaches the crystalloid condition, the greater is the concentration of salt required for salting out. Thus, for instance, the alhumoses are classified in accordance with concentration of salt required for their precipitation (see p. 166.) As early as 1907, Bechhold had already called attention to the relation between salting out, and particle size and salt concentration. S. Oden produced the experimental evidence that reversible sul- phur and silver hydrosols could actually be separated in fractions, OPTICAL AND ELECTRICAL PROPERTIES OF COLLOIDS 81 distinguishable by the size of their particles, Ijy "fractional coagula- tion/' ^ i.e., by the addition of salt solutions of progressively increas- ing strength. The influence of neutral salts exhibits regularities which are of great importance in a number of biological phenomena and which we shall repeatedly encounter. The salting out of hydrophile colloids, gelatinization, the irritability of muscle and nerve, the permeability of cell membranes (blood corpuscles, etc.), the swelling of membranes and many other phenomena are thus related to a group of physico- chemical properties of solutions, whose relationship is indubitable even though the true basis is not clear. With H. Freundlich we shall describe these effects of neutral salts as lyotropic (solution changing), and we shall study them more closely. Most inorganic salts increase the surface tension of water and from a table of W. K. Röntgen and J. Schneider we obtain the following series for the increase in the surface tension by the alkaline iodids: Na > K > Li > NH4. With the anion of various alkalis: CO3 > SO4 > CI > NO3 > I. Though the compressibility, the solubility and the viscosity of water are changed in a similar order by neutral salts, the relation- ship is much more fundamental. Neutral salts may accelerate or impede catalytic effects, such as the inversion of cane sugar, the saponification of esters and the changing of acetone into diacetone alcohol. The action in acid solutions is usually the reverse of what it is in alkaline solutions as has been shown by R. Höber. In acid media the acceleration by cations is Li > Na > K > Rb > Cs, in alkaline media Cs > Rb > K > Na > Li. For anions in acid solutions the following order holds: I > NO3 > Br > CI > CH3CO2 > SO4. In alkaline solution the series is reversed. In neutral solution also the lyotropic series holds, although small changes in the arrangement may exist for some of the ions. ^ Since reversible hydrosols are considered I might describe the procedure preferably as "fractional salting out." It is particularly interesting to know; that according to Oden and Oholm the particles do not coalesce but retain their identity and when they are redissolved, there are as many particles in solution as before the salting out. 82 COLLOIDS IN BIOLOGY AND MEDICINE We encounter such lyotropic series regularly in the salting out of hydrophil ecolloids (see Albumin, p. 146 et seq., Lecithin, p. 140, Gelatin, p. 161) and in many biological phenomena, where they tend to cause either a precipitation or a concentration. However, we must not assume that the action is such that in one case the anion alone is active, and in the other the cation alone. There are many reasons for believing that there is an antagonism between anions and cations, and that the action of the cation is more or less powerfully increased or diminished depending upon the anions present. If in a given instance we speak of the cations causing precipitation or dehydration, we always mean the difference between the effect of the cation and the opposed action of the anion present, in which, however, the action of the cation predominates. The divalent cations Mg, Ca, Sr and Ba act more strongly as pre- cipitants than the monovalent cations. Biologically they are im- portant in connection with alkali salts, inasmuch as small quantities of calcium salts are able to replace considerably larger quantities of alkali salts, e.g., Na or K. This greater effect may even lead to an antagonism between the two; it has been thoroughly studied by Wo. Pauli and H. Handovsky.*^ It is sufficient to mention that in the case of alkali albumin, Ca salts form a less ionized Ca albumin combination from a more ionized Na albumin combi- nation. According to the law of mass action small quantities of Ca may be replaced by larger quantities of Na. We may thus under- stand the significance of the Ca content in all physiological fluids. Mg, Sr and Ba act in a similar manner. They have in addition certain specific properties which obscure the relations. Large quantities of alkaline earths cause irreversible changes in many biocolloids, that is, they produce insoluble compounds with them. Electrolytes may exert another effect on the biocolloids; they may cause "flocculation," a phenomenon we shall now study more closely. Flocculation. If albumin is boiled, it coagulates. A coagulation of albumin may also be produced by the addition of ammonium sulphate. Though the latter process may be reversed by dilution with water, the boiled albumin cannot again be brought to its fluid condition, by any physical means. The hydrophile colloid has been converted into a hydrophobe colloid. For uniformity's sake, we shall consider as co- agulations only such processes as cause an irreversible change. If we heat a very dilute albumin solution, there is apparently no coagulation; at most the fluid becomes slightly opalescent. In OPTICAL AND ELECTRICAL PROPERTIES OF COLLOIDS 83 reality a coagulation has occurred here, since the addition of a drop of acetic acid or a little ammonium sulphate produces flocks. The process is therefore called flocculation. These albumin flocks are in- soluble in water. Only by the second process, by the agglomeration of the smallest particles (see Plate I, h and c) under the influence of the acetic acid or ammonium sulphate was it possible to separate the two phases (water/albumin). This occurs in a way similar to that in which the drops of rain are formed under certain conditions by the union of particles of mist. The union is preceded by a slowing of the Brownian-Zsigmondy movement, as has been shown with the ultra- microscope by V. Henri.* Flocculation is an electrical 'phenomenon. It is brought about by electrolytes as well as hy colloids of opposite electric charge, as well as by ultraviolet and Roentgen rays (the action of rays is much weaker than the action of electrolytes).^ If we shake purified lampblack with water we get a suspension which remains turbid for weeks. If we pour a few drops of an alcoholic mastic solution into water, it remains milky for weeks and even years. We have seen that the hydrophobe inorganic hydrosols, such as colloidal arsenic sulphid, platinum sol prepared according to Bredig's method, gold sol pre- pared according to R. Zsigmondy's method, are permanent for months provided the solution is free from electrolytes. Addition of electrolytes causes an irreversible flocculation of these hydrophobe colloids. The process is to be sharply distinguished from salting out, i.e., the reversible precipitation of albumin, albumoses, etc., from solution by large quantities of salt. The phenomenon of flocculation is encountered especially in the precipitation reactions of albumins and in Chapter XIII on Im- munity Reactions, where it plays an important role in the agglutina- tion of bacteria by precipitins. Moreover the precipitation of gold hydrosols by cerebrospinal fluid has acquired great significance in the diagnosis of mental diseases (see p. 354). To cause flocculation, a certain minimum amount of electrolyte as as well as of the dispersed phase is required. Below these limits, which are characteristic for every electrolyte, no flocculation occurs even after months. H. Bechhold called this minimum, the "elec- trolyte threshold" and the "suspension threshold.'' The rate of flocculation is dependent on the concentration of the suspension and of the electrolytes, i.e., the more concentrated the sus- "■ Specific immunity reactions, in which precipitations occur, are probably not electrical (see p. 197). 84 COLLOIDS IN BIOLOGY AND MEDICINE pension and the electrolytes within certain limits, the greater is the rapidity of flocculation. The dependence upon the concentration of the electrolytes is especially noticeable in the vicinity of the elec- trolyte threshold. Further away from the electrolyte threshold the rapidity of flocculation is less dependent upon the concentration of the electrolyte. (H. Bechhold.*^) Under certain conditions, an excess of electrolytes may lead to re-solution. This phenomenon is called peptisation. Graham first observed this occurrence upon treating ferric oxid gel with ferric chlorid and compared it to the formation of water-soluble peptone from coagulated albumin and hydrochloric acid. As a matter of fact, peptisation probably depends upon a renewal of electric charge by the excess of electrolytes. As has been shown by H. Freundlich and his pupils Ishizaka and ScHUCHT *^, these facts furnish an indication that the more sen- sitive a substance is to flocculation, the more will it be adsorbed. The great majority of colloids migrate to the anode, and on floccu- lation the cation is of much the greatest importance; the anion plays but a subordinate role. The conditions are reversed in the case of the few colloids which migrate to the cathode. However, R. Burton has shown that with increase of electrolyte concentration the rate of migration of the colloid becomes constantly diminished and finally the direction may change. At the stage when the concentration of electrolytes is such that reversal occurs, in other words the isoelectric zone, flocculation is most rapid. The increase in the action of the cations is out of proportion to the increase in their valence. The electrolyte concentration necessary for the flocculation of a mastic suspension is FeCls : BaCl2 : NaCl = 1 : 50 : 1000. There are also certain anomalies (see H. Bechhold,*i as well as M. Neisser and U. Friedemann*) in connection with the rate of migration of the ions, as well as in the electrolytic dissociation and especially in the ionization pressure of electrolytes. In general the above relation between the action of the cations is maintained. The powerful flocculating action of the H ion without doubt depends upon its high speed of migration while the OH ion has a corresponding action upon colloids which migrate to the anode. Flocculation by means of trivalent iron and aluminium salts show peculiar anomalies. These were discovered by M. Neisser and U. Friedemann* and also by H. Bechhold*^ and called irregular series. Later these phenomena of Inhibition Zones were followed further by 0. Teague and B. H. Buxton*^ and also by A. Lottermoser.* An example from my published paper will best explain the phenom- enon. XXX means strong flocculation, X X medium, X none. OPTICAL AND ELECTRICAL PROPERTIES OF COLLOIDS 85 Concentration of mastic 0.01 0.005 0.0C25 0.001 0.0005 0.00025 solution. 1/2 XXX XXX XXX XX XXX XX 1/4 .XXX XXX XX XX XXX 0 1/8 XXX XXX 0 0 XXX X 1/16 XXX XX 0 0 XX XXX 1/32 XXX 0 0 0 0 XXX Mol. Al2(SO^)3 The discoverers explain the phenomenon by saying that the salts mentioned undergo strong hydrolytic dissociation so that the col- loidal iron and aluminium hydroxid present in the solution act as ''protective colloids" somewhat like gelatin or albumin. The flocculation hitherto described occurs only with true hydro- phobe colloids. If the individual particles possess a film of native albumin, gelatin, an albumose (such as Na lysalbinate), dextrin, etc., the particles act as if they were the "protective colloids" involved (see p. 77); in other words they are salted out only by very large amounts of electrolytes. The colloid precipitate can be redissolved in water providing the protective colloid is not denatured by the electrolytes. On this account all the colloidal metal sols used thera- peutically such as collargol, bismon, etc., are "stabilized" by hy- drophile colloids. Without this protection it would be impossible to preserve them for any length of time and they would be floccu- lated when they were prepared for intravenous use by dilution with physiological salt solution. R. ZsiGMONDY considers as characteristic of hydrophile colloids, the protection they give to his gold hydrosol against flocculation by electrolytes. He designates as the gold figure, the amount (in mg.) of the colloid in question, which was just sufficient to prevent the 2 flocculation of 10 c.c. of a gold sol. by 1 c.c. - NaCl sol. The floccu- lation is accompanied by a change in color from red to blue, and it is, sufficient to observe this change. The following data from R. Zsigmondy illustrate this point: Colloid. Gold figure in mg. Gelatin Casein. . . 0.005-0.01 0 01 Egg albumen Gum arable 0.06-0.3 (depending upon its origin and ash content) 0.15-0.25 Tragacanth Dextrin 2± 10-20 Potato starch Sodium stearate .... Sodium stearate .... Sodium oleate 25 10 (added at 60°) 0 . 001 (added boiling hot) 0.4-1 86 COLLOIDS IN BIOLOGY AND MEDICINE Peptones have no protective action at all [Peptones may cause precipitation. E. Zunz, Bull. Soc. Roy. des Sc, Med., et Nat., June 11, 1906. Tr.], whereas some of the albumoses, especially sodium lysalbinate and sodium protalbinate have a very powerful protective action which C. Paal utilized in preparing a large number of inor- ganic colloids. Sensitiveness to flocculation may vary considerably with the nature of the protective colloid. The flocculation of hydrosols may be brought about not only by electrolytes but also by hydrosols, providing they have an opposite electric charge, as has been shown by W. Biltz. Thus, for instance, arsenic trisulphid, gold and platinum hydrosols, etc., are flocculated by ironoxid, aluminiumoxid, chromiumoxid hydrosols, etc. A proper relative mixture is required, which means the charge of the positive sol must be counterbalanced by the charge of the negative sol. If a sol is in excess, no flocculation occurs, and the entire complex con- sisting of both colloids migrates, when placed between two elec- trodes, in the direction of the sol (J. Billiter*) which is in excess. This explains why protective colloids may under some circumstances produce flocculation instead of protection, namely, when they are added in minimal quantities. Thus, for instance, mastic emulsions are flocculated by 0.0003 to 0.0001 per cent of gelatin (H. Bechhold and also M. Neisser and U. Friedemann). Hydrochloric acid in a dilution incapable of producing flocculation by itself can coagulate gold hydrosol, mastic or oil emulsion in the presence of one part of gelatin per million. As has already been mentioned we must not confuse the salting out of hydrophile colloids with flocculation, though there are border- line cases which complicate the phenomenon. If we add for instance a salt of a heavy metal to a dilute albumin solution, a more or less irreversible albumin-metal compound will form, dependent on the nature of the salt upon which the excess of albumin acts as a pro- tective colloid. Acids may cause a loss of charge and the metal salt may then exhibit some flocculating and some salting out action. Such a case might occur, for instance, on adding zinc sulphate to albumin (studied by Wo. Pauli). The process is as follows: ZnS04 + Albumin. 0.05 n maximum flocculation (irreversible till 1 n). In precipitate disappears. 2 n precipitate reappears (reversible) . 4 n maximum precipitate (reversible). There are transitions between hydrophobe and hydrophile col- loids which are responsible for the transitions between flocculation OPTICAL AND ELECTRICAL PROPERTIES OF COLLOIDS 87 and salting out. Cholesterin and lecithin may be regarded as such transitional substances, whose flocculation has been thoroughly studied by 0. Porges and E. Neubauer.* Cholesterin closely ap- proaches the hydrophobe colloids, lecithin the hydrophile, and on this account the former is irreversibly precipitated by salts and the latter reversibly. Yet the concentration of alkaline earths required to precipitate lecithin is considerably less than for the more hydro- phile albumin, and conversely much greater concentration of salt is required for the flocculation of Cholesterin than is the case with true hydrophobe sols. In the case of lecithin, the "irregular series" occurs even with neutral salts (magnesium and ammonium sulphate). As will be seen in Chapter XIII, such transitions from hydrophobe suspensions to hydrophile colloids may be artificially produced with emulsions of bacteria. Much has been written on the theory of salting out and flocculation. No one theory accounts for all the individual facts, yet the following explanation is generally useful. Flocculation is brought about by the coming together of small particles to form larger complexes. These agglomerations always occur under the influence of electric forces and in fact the optimum for reversible salting out is in the iso- electric zone (see p. 158). The process, must, therefore, be brought about by a discharge of the particles. Risdale Ellis has shown by researches on oil emulsions, that the charge at the interface between water and the dispersed phase is probably reduced to a minimum by the addition of precipitating electrolytes. The smaller this charge is at the interfaces (electric double layer), the more readily the double layer is broken down, resulting in a union of the suspended fluid or solid particles. In the case of anodic colloids, cations of an electrolyte, and in the case of cathodic colloids, anions or an oppositely charged colloid, lessen or neutralize the electric charge, so that the particles may unite. (See H. Freundlich's " Kapiflarchemie " and Wo. Ostwald's "Grundriss der Kolloidchemie.") Irreversible precipitation of metal hydrosols frequently occurs out- side the isoelectric zone and the electrolyte threshold is not as sharp as with reversible hydrosols. Radioactive Substances as Colloids. It has been demonstrated by electric migration and dialysis that radioactive substances occur in colloidal solution. In view of the great biological significance of such substance we shall explain the facts more fully. 88 COLLOIDS IN BIOLOGY AND MEDICINE GoDLEWSKi * found upon electrolysis of the radium emanations and their products that the radioactive substances migrated mostly to the cathode in acid solutions and to the anode in alkaline solutions. Paneth dialyzed a solution of radio-lead nitrate against pure water in a parchment thimble and greatly concentrated the RaE and polonium in the thimble but the relation between RaD to lead was not changed. Both phenomena are characteristic of colloids and indicate the colloid properties of radioactive substances; this led them to continue their investigations. Godlewski * successfully adsorbed RaA, RaB and RaC with inorganic colloids and concen- trated the radioactive substances by precipitating the latter. He was successful not only with radium but with actinium, mesothorium and uranium. The concentration of radioactive substances by adsorption on colloidal silicic acid (Ebler, Fellner) has attained great practical value in the manufacture of radium preparations. In fact RaA and RaC may be collected from acid solutions simply on filter paper and by burning the paper a highly active ash, free from RaB, may be obtained. It is evident that in solution radioactive elements undergo hydrolysis with the formation of a colloidal radio- hydrosol. CHAPTER VII. METHODS OF COLLOID RESEARCH. A FIELD of research extending as does colloid chemistry into so many other branches of science must be served by countless methods. Purely chemical as well as physical and biological methods which the investigator of colloids utilizes in his studies, have been so well developed and so thoroughly described in technical literature, that it is needless to discuss them more fully here. There are, however, several methods which are peculiar to colloid investigation, and we shall consider them here. Unfortunately, the limits of their usefulness are as yet unestablished; for some one to establish them would be highly desirable. In discussing the following methods, I have not sought complete- ness but have considered only those which have proved practical; I have personally tested most of them. To determine whether a solution of a substance is colloidal in character it must be tested by dialysis, ultrafiltration or diffusion. Dialysis is a purely qualitative method which determines whether or not a substance is colloidal in character, i.e., whether or not it consists of large particles. Ultrafiltration in most cases may be used instead of dialysis; it works much more rapidly and above all permits, in addition, quan- titative experiments, and consequently has a much broader utility. Both methods serve to separate colloids from crystalloids. This separation occurs with dialysis if the water surrounding the dialyzer is frequently renewed; it occurs with ultrafiltration, provided the substance on the funnel is washed repeatedly, as in an ordinary filter. Diffusion is an excellent quantitative method for the investigation of particle size. The performance of diffusion experiments is, how- ever, a difficult matter, because even variations in temperature may cause material errors. Dialysis.^ The most varied apparatus and membranes may be used for dialysis. An apparatus described in most textbooks and used in many ^ An exhaustive description of all known methods of dialysis is given by E. ZuNZ in Abderhalden's Handb. d. biochem. Arbeitsmethoden 3, pages 165-189 and Supplement, pages 478-485. 89 90 COLLOIDS IN BIOLOGY AND MEDICINE teaching laboratories is the one originally described by Graham. A wide-mouthed salt bottle A (Fig. 9), with its bottom broken ojff, has a pig or ox bladder or a piece of parchment bound about the neck; the bottle is placed with the membrane downward, in a vessel of water {By. The solution to be dialyzed is placed in the bottle. Precautions: Before the membrane is used, it should be tested to see that it can be made wet by water. Greasy animal membranes must be rinsed with fresh water several times, on both sides. To determine whether the Fig. 9. A simple dialyzer. membrane leaks, a colored solution {e.g., a drop of colloidal silver, Utmus, or water colored with hemoglobin) is placed in A and allowed to remain there several hours, without putting any water in the outer vessel. Colored drops will pass through at points of leakage (the margin where the membrane is bound should be espe- cially watched) . To make sure that the drops are really colored solution and not pure water, they should be absorbed by filter paper. In all dialysis experiments, the substance under examination may simulate colloidal character by being bound or adsorbed by the dialyzing membrane. Under these circumstances, if we wish to determine whether the solution under examination contains colloidal substances, it is necessary to use the smallest possible membrane with the largest available amount of substance. If too little substance is used, it may all become bound by the membrane, and in spite of the fact that it is not colloidal, none will be found in the dialyzer. But if so much of the substance under investigation is used, that in spite of any possible ad- sorption by the membrane, plenty still remains in the dialyzer, then an examina- tion of the water outside wiU settle the question. In practice, Graham's dialyzing apparatus is not frequently em- ployed, because it has a small dialyzing surface, and this is a great disadvantage. In order to bring the largest possible surface into contact with the surrounding water, there are used either whole pig bladders, ox bladders, fish bladders, or the commercial parchment thimbles.^ I have had very good results with fish bladders (condoms) which are very thin, uniform and elastic, though unfortunately they are expensive. Parchment thimbles, referred to above, are recommended for the dialysis of large quantities of solution. They may be obtained in all sizes and in all lengths. The suspension of a fish bladder is conveniently accomplished by pressing it between two glass rods which are held together by rubber bands (cut from gas tubing) ; the 1 With organic solvents, instead of water, alcohol, benzol, etc., must of course be employed and the membrane (preferably collodion) is previously soaked in these fluids. ^ These are on sale at the "Vereinigten Fabriken für Laboratoriums-bedarf" Berlin, Scharnhorst Str. (The Kny Scherer Co., N. Y., are the American agents.) METHODS OF COLLOID RESEARCH 91 Fig. 10. A fish bladder (condom) dialyzer. glass rods are laid over a tall narrow beaker, in which the water is placed. Parchment tubes are suspended in a similar manner. To avoid tying one end of the tube, it is hung up in U-fashion so that both open ends are pressed together by the glass rods (see Fig. 10). Precautions: In filling and hanging the parchment tubes the air must be en- tirely pressed out before fastening the ends, for should water dialyze in, consider- able pressure will develop, which may burst the mem- brane. All the precautions given on page 90 must be observed (degreasing, absorption, etc.). Excellent dialysis membranes may be pre- pared of collodion or glacial acetic acid-collo- dion. Their great advantage is that they may be prepared in any desired size, shape and degree of permeability, and are easily sterilized. As an example, we shall explain how to make one such membrane. A test tube is dipped into collodion and then allowed to drip, being twirled meanwhile, and when it has skinned over, the whole tube is quickly immersed in water. In a short time, the tube is circumcised at a height which will give a membrane of the desired length; with a little practice, the membrane can be removed from the tube with ease. The membrane may be formed inside the tube also by rinsing the test tube with collodion or glacial acetic collodion and then filling it with water. In a similar manner, spher- ical or cylindrical membranes of 30 to 40 cm. long and 10 cm. diameter may be made (see Fig. 11). Such sacs may be fastened to a glass tube with thread or collodion so as to form a water- tight joint. The membranes are best preserved in water to which a little chloroform has been added to prevent the growth of moulds. In making his collodion sacs, G. Malfitano ^ uses glass tubes of the shape shown in the accompanying diagram, which are rotated by a motor to secure uniform drying. The spherical swelling affords an easier removal of the rim (see Fig. 12). After the cut is made through the collodion skin at the equator of the sphere (— »), it is carefully loosened from the glass and turned inside out or the rim is fastened to a large glass tube which is exhausted, thus the skin is loosened from the spherical portion. J. Duclaux uses very thin tubes, about 1 cm. in diameter and 1 meter long, so as to get the largest possible surface. ^ According to a personal communication. 92 COLLOIDS IN BIOLOGY AND MEDICINE G. Malfitano and J. Duclaux use their sacs chiefly for ultrafil- tration though they are equally useful for dialysis. Biologists fre- quently use little reed sacks for dialyzing; they are frail though sterilizable, but their capacity is small. Schleicher and Schüll (Düren) sell dialyzing thimbles. They are tubular, closed at the bottom, moderately firm, made of parch- FiG. 11. Collodion sacs. (A. Schoef.) ment and, since they maintain their form, they are useful for certain experiments. They are employed in the Abderhalden test. R. ZsiGMONDY *=* constructed a very useful "star dialyzer." ^ The hard rubber ring B (see Fig. 13) is covered by a membrane (col- lodion, parchment or the like) and is placed on a plate A, which carries the star-shaped support. Through the central opening in the ^ Obtainable from Robert Mittelbach, Göttingen. METHODS OF COLLOID RESEARCH 93 plate, water flows and bathes the extensive dialyzing surface of the dialyzer A. JoRDis has constructed an apparatus resembhng a filter press for dialyzing large quantities, A number of wooden rings are soaked in /Ai ^B^A Fig. 12. Tube for the preparation of collodion sacs. (G. Malfitano.) (R. 13. Star dialyzer. Zsigmondy.) paraffin and then both surfaces are covered with parchment. These rings are placed in an apparatus G so that between each ring with parchment there is one without parchment. They are made tight with rubber washers. The individual elements of the filter are iVooc/en Ring, I cm. thick ^^ W= Running Water 0=Diaiysing Spaces Fig. 13a. Continuous dialyzing apparatus of E. Jordiss. Reproduced from Zeitschrift fur Electrochemie, p. 677, Vol. VIII, 1902. pressed together with wing nuts so that they are water-tight. The solution to be dialyzed is placed in the rings covered with parch- ment, the water in the intervening rings circulates through holes made in them for the purpose, see Fig. 13a. 94 COLLOIDS IN BIOLOGY AND MEDICINE An excellent apparatus ^ for continuous dialysis which also permits a concentration of the dialyzate has been constructed by Kopac- ZEWSKi. He apphes the idea of the Soxhlet extraction apparatus. A collodion sac prepared as described on page 91 is placed in the tube A filled with water. The dialyzate from the collodion sac may be removed either through the cock C or may pass either at once or drop by drop into the vessel B. If B is heated, the water vaporizes and /^b a S. KOPACZEWSKfS APPARATUS Fig. 13b. Fig. 14. Dialyzing filter. (L. Morochowetz.) passes into the two condensers and drops again through a and 6 into the tube containing the collodion sac. Since with biological fluids the temperature must remain below 50° C, the heating is done under partial vacuum (reduced pressure). The tube is emptied through the lateral outlet having a cock c. A concentrated dialyzate is finally obtained in the vessel B. Where there is no running water available for dialyzing, the dialy- zing filter of Leo Morochowetz may be employed. Its arrange- ment may be seen in Fig. 14. The funnels may be obtained from any supply house and the parchment filters from Schleicher and ScHULL (Düren). 1 The apparatus may be obtained from Poulenc freres, 122 Boulevard St. Germain, Paris, France. METHODS OF COLLOID RESEARCH 95 Dialysis is considerably hastened by agitating the dialyzing fluid. I have never found it suggested that the contents of the dialyzer should be stirred but this is a useful procedure with non-foaming solutions. F. Hofme-ister fastens all the dialyzing thimbles to a common rod which he rocks up and down with a motor. R. Köhler places fish bladders in wide-mouthed bottles which he closes with a rubber cork and a rubber cap, and then places it in a shak- ing machine; to prevent twisting off the fish bladder at its neck, he inserts sev- eral glass rods as shown in Fig. 14a. Ultrafiltration. PUBBE/? CAP RUBBER STOPPER -FISH BLADDER ■GLASS RODS Fig. 14a. H. Bechhold defines ultrafiltration as filtration through jelly filters. They serve to separate colloid solutions from crystalloids and for the separation of colloid mixtures having parti- cles of different size. If we know the size of the pores in an ultra- filter, ultrafiltration affords information as to the size of the particles in the colloid under investigation. Ultrafilter. For ultrafiltration, sac-like membranes may be em- ployed prepared as for diffusion experiments (see p. 91) and mounted as shown in Fig. 11. A. ScHOEP * increased the permeability of membranes by adding glycerin and castor oil to the collodion. This is of great importance in filtering inorganic colloids. This sort of ultrafiltration is used especially in France (G. Mal- FiTANO, J. DucLAUx), but it is of limited utility. Filtration occurs very slowly (a few cubic centimeters per hour) and the filters can- not withstand much pressure, so that their usefulness is very limited. For ultrafiltration, H. Bechhold** used pieces of filter paper im- pregnated with jellies. By means of this paper support the filters acquire great strength and may at times sustain in Bechhold's ultra- filtration apparatus, pressures of 20 atmospheres or more. Since H. Bechhold discovered that the premeability or tightness of the ultrafilters depended on the concentration of the jellies used in pre- paring them, it is possible to make filters with pores of any desired size. The filters may be purchased ready-made.^ 1 Schleicher and SchüU, in Düren, market Bechhold's ultrafilters in aluminium boxes which contain 10 filters filled with water and sealed with a rubber ring, (diam. 9 cm.). This firm keeps in stock six kinds, of different porosity. 96 COLLOIDS IN BIOLOGY AND MEDICINE Since some may desire to make the filters, brief direction for doing so are given. ^ The most useful kinds of filter paper are No. 566 and No. 575 of Schleicher and Schüll, These are cut into discs of 9 cm. in diameter and impregnated with the jellies under atmospheric pres- sure in a glass trough from which the air has previously been ex- hausted, making a vacuum. ^ The square trough T (see Fig. 15) has its cover ground airtight. On the cross bar B a number of filter papers are suspended. The cover C has two openings; through No. 1 pass two tubes, one of which leads to the air pump A and the other to the pressure gauge Fia. 15. Trough for the preparation of ultrafilters. (H. Bechhold.) m. When the air is exhausted from the trough, the fluid jelly is al- lowed to enter through the funnel F, which has a cock and a tube leading to the bottom, until sufficient is admitted to cover the filters. Then the valve leading from the funnel is closed and the valve through which the air was exhausted is opened so that the jelly is forced into the filters under atmospheric pressure. After a time (with diluted jellies 10 to 20 min. with concentrated jellies one or two hours) the cover is taken off and the rod with the filters is re- moved from the fluids. While the filters are draining, they are constantly shaken. Finally the v/hole filter is rapidly gelatinized by plunging it into a suitable fluid. In the case of glacial acetic acid 1 Given in detail in the original paper of H. Bechhold.** 2 May be obtained from the Vereinigten Fabriken für Laboratoriumsbedarf, Berlin. METHODS OF COLLOID RESEARCH 97 collodion, water is used. If gelatin is being used, the entire impreg- nation trough must be placed in a bath of lukewarm water. Gelatin filters are hardened by placing the filters, still moist and gelatinized in the air, into a 2 to 4 per cent ice cold formaldehyde solution and keeping them for a time in an ice box. The filters, however prepared, are washed several days in running water and preserved in water to which a little chloroform has been added to prevent the growth of mould. H. Bechhold generally uses glacial acetic acid collodion, a solu- tion of soluble cotton in glacial acetic acid.^ By diluting with glacial acetic acid the solution may be reduced to the desired concentration. If non-aqueous solutions {e.g., benzol, ether, etc.) are to be ultra- filtered, the water must be displaced by a series of solvents. (Water is first displaced by acetone, the latter by benzol, and so on.) Fig. 16. Ultrafilter. (H. Bechhold.) The Ultrafiltration Apparatus. Very porous filters are permeable under low pressure and can be used like any other filter. In by far the largest number of cases, a pressure of from 1 to 20 atmospheres must be exerted to obtain any filtrate at all. For this purpose H. Bech- hold prepared an apparatus which in its simplest form is shown in Fig. 16 2; Fig. 17 is more suitable for very high pressures. Fig. 16 consists of a cylindrical vessel H into which is inserted the funnel T. Between the lower flanges of T and H, the disc of filter paper is pressed. This is made tight by the two rubber rings G, G. To protect it from being torn, the filter lies on a nickel netting or perforated nickeled plate N and is further protected from bulging un- der pressure by the plate P, which has several holes in it. The upper part of the funnel T is ground conically and is closed by means of ^ The Chemischer Fabrik auf Aktien (vorm. Schering), Berlin, prepares to order solutions of 10 per cent collodion and 2\ per cent potassium carbonate, which show only slight tendency to contract when they gelatinize. 2 All this apparatus is manufactured by the Vereinigten Fabriken für Labora- toriumsbedarf, Berlin, Scharnhorst Str. 98 COLLOIDS IN BIOLOGY AND MEDICINE Fig. 17. Ultrafilter for high pressures. the cover D with a conical joint and a rubber washer. By turn- ing the screw cap Sehr, the cover above as well as the filter below is tightened. A small nipple with a screw thread passes through the cover and to this the pipe from the pressure chamber is attached. The apparatus shown in Fig. 17 is chiefly used for pressures above 10 atmospheres, and is closed with flanges. Naturally this is more bulky. The lettering in Fig. 17 corresponds with that of Fig. 16, so that it is unnecessary to duplicate the description. An apparatus with a stirrer (also on the market) is usually prefer- able, because filtration is relatively more rapid and the filtrate is more uniform. The omission of stirring may permit a gel layer to form on the ultrafilter, and this gel layer may then act as a filter itself. It is especially important to have the packing tight against high pressures. In this apparatus the pressure is introduced through a side opening because the stirrer occupies the central one. The Pressure. The pressure may be produced by a hand pump. This is especially useful in the scientific investigation of the action of filters, where fine gradations of pressure are involved, and where prolonged pressure is unnecessary. In practical ultrafiltra- tion, it is preferable to use a steel cylinder containing either com- pressed air, nitrogen or carbonic acid, etc. Between the steel cylinder of the ultrafiltration apparatus, a reducing pressure valve and two manometers are introduced; one for very high pressure shows the pressure in the cylinder, the other, beyond the valve, the lower pressure in the ultrafiltration apparatus. A second reducing pressure valve near this manometer permits such delicate differences in pressure that, in my opinion, it is safe to use this arrangement in- stead of the hand air pump even in scientific measurements. The extensive use achieved by Bechhold's ultrafiltration has led to some modifications for special purposes. Zsigmondy, Wilke- DÖRFURT and Galecki recommended collodion skins for analytical purposes. They placed collodion skins on a Büchner funnel (Gooch filter, i.e., porcelain funnel with perforated diaphragm) and thus filtered off coarse colloids, especially inorganic ones, by means of a tap-water suction pump. This arrangement is not suitable for pressures above one atmosphere. [J. F. McClendon employs alun- dun thimbles coated with collodion. Tr.] The essential new point in the apparatus devised by Burian E. METHODS OF COLLOID RESEARCH 99 Pribram and Kirschbaum is the application of compressed air for stirring; the gas which supplies the pressure enters the bottom of the vessel through a perforated spiral and thus agitates the fluid. I have had no occasion to determine whether this has any advantages over the mechanical stirrer. The Gauging of Ultrafilters. It is important in many cases to have a measure for the limits of effectiveness of ultrafilters, as in this way we may obtain information concerning the size of the par- ticles of the colloid under investigation. The following methods given by Bechhold are suitable for the purpose: 1. Hemoglobin Method: A 1 per cent solution of hemoglobin (hemoglobin scales, Merck) is prepared and the filter is tested to see if it permits the hemoglobin to pass through. If the hemoglobin is retained, the filter is impermeable to most inorganic colloids (with the exception of freshly prepared silicic acid). The degree of permeability of the filter for hemoglobin may be recognized by the intensity of the red color in the filtrate. H. Bechhold has prepared the following table of permeability for ultrafilters, which is arranged in the order of the diminishing size of the particles of the colloids in solution, and was obtained by using ultrafilters having different degrees of porosity. Suspensions. Prussian blue. Platinum-sol, Bredig. Ferric oxid hydrosol. Casein, in milk. Arsenic sulphid hydrosol. Gold solution, Zsigmondy, No. 4, c. 40 fxjjL. Bismon, colloidal bismuth oxid, Paal. Lysargin, colloidal silver, Paal. Collargol, silver, v. Heyden, 20 /x/i. Gold, solution, Zsigmondy, No. 0, c. 1-4 iJLfjL. 1 per cent gelatin solution. 1 per cent hemoglobin solution, molecular weight c. 16,000. Serum, albumin, molecular weight 5000 to 15,000. Diphtheria toxin. Protalbumoses. Colloidal silicic acid. Lysalbinic acid. Deutero albumoses A. Deutero albumoses B, mol. wt. c. 2400. Deutero albumoses C. Litmus. Dextrin, mol. wt. c. 965. Crystalloids. 2. Air Transpiration Method.^ This method affords approxi- mately absolute values for the largest pores of an ultrafilter. It is based on the following principle. In order to force air through a 1 Before actually undertaking methods 2 and 3 the original paper should be consulted (Bechhold*^), as the details cannot be abstracted. 100 COLLOIDS IN BIOLOGY AND MEDICINE capillary completely immersed in and wet with water, a certain pressure is necessary, which depends upon the surface tension of water against air (which is a constant) and the diameter of the capillary. If D is the diameter of the capillary, p the pressure in atmospheres and ß the capillarity constant, the following formula applies : D= ^ p.1.033 + 10' li ß = 7.7 at 18°, we obtain D = 30.8 OP' p.1.033 + 10^ With the aid of this formula, the smallest diameter of the pores in question may be calculated from the least pressure necessary to drive air through the pores of the completely wet filter. The practical performance of the experiment is as follows: The filtering apparatus is turned upside down, a thin layer of water is placed on the filter (several milli- meters high) and the highest pressure at which air bubbles begin to escape is determined. Casing \ " The diagrammatic sketch, Fig. 18, shows the filtering apparatus in normal position {T = funnel, F = ultrafilter, A = air intake). Fig. 19 shows it in the position necessary for the forcing through of air. Above the filter there is a thin layer of water. According to this method the largest pores of a filter which just permits hemoglobin to pass through have a diameter 50 to 99 ix/x. 3. Method Based upon the Rate of Transfusion for Water. This method affords approximately absolute values for the average diam- eter of the pores of ultrafilters. The method is based upon the somewhat indefinite law of Poiseuille for the passage of fluids through capillary tubes.^ D = diameter of the pores. Q = amount of water flowing through the surface F under the constant pressure S. R is the ratio between the empty (water containing) space, and that filled with solid. This is determined from the percentage of dry material in the jellies (a 5 per cent filter contains full and empty space in the proportion of 5 to 95). 1 Bechhold.*« METHODS OF COLLOID RESEARCH 101 L is the length of the capillaries {i.e., not smaller than the thick- ness of the wet filter). Ä; is a constant factor dependent on temperature and kind of fluid. The following formula applies : Q{R + I)L D = k'S'F-R If all the experiments are performed under the same conditions, the formula may be simplified, because ^ — ^ — ^ becomes a constant. For the practical performance of this experiment two persons are required, one of whom regulates the pressure, while the other de- termines the amount of water filtered over a certain time, fixed by means of a stop watch. Under the apparatus is placed a funnel which has a rubber tube with a pinch-cock attached. The ultrafiltration apparatus is filled with water, and the air pres- sure is raised to a given point. The pinch-cock is then closed so that all the water filtering at a constant pressure is caught in the funnel. After a given time {e.g., one minute) has elapsed, the entire pressure is instantly released. In this way the amount of water filtered through a given filter in a given time is measured. If we have previously performed the same experiment with a filter paper which has pores of known size, one which, for instance, even par- tially retains blood corpuscles or bacteria, i.e., objects which are measurable microscopically, by means of the above formula, we can estimate the average size of the pores of the ultrafilter. With this method ultrafilters which just held back hemoglobin showed the average diameter of their pores to be from 33 to 36 fx/j,. 4. Method of Emulsion Filtration, described on pp. 15 and 16. Adsorption by Filters. In ultrafiltration experiments, it is necessary to avoid errors due to adsorption on the part of the filter. Accordingly, as a preliminary experiment, it is advisable to shake a portion of the solution with a shredded filter. If the content of the solution is practically the same afrte as before the shak- ing, there has been no adsorption. If the adsorption introduces an error into the ultrafiltration experiment, it is necessary to use a dif- ferent jelly. For instance, arachnolysin is very strongly adsorbed by glacial acetic acid-collodion, but only slightly by formol- gelatin. In ultrafiltration experiments it is always important to work quantitatively and to test what remains in the filter as well as the filtrate obtained. 102 COLLOIDS IN BIOLOGY AND MEDICINE [P, A. KoBER has devised a new and valuable form cf ultiafilter based on the principle of selective dialysis through collodion and evaporation of the dialysate (per-vaporation). See Jour. Am. Chem. Soc, Vol. XL, No. 8, po 1226, et seq. Tr.] Applications of Ultrafiltration, ultrafiltration as previously men- tioned serves to separate colloids from crystalloids. It can fre- quently replace dialysis, having the advantage of rapidity and permitting separation without the unavoidably great dilution of the dialysate. For this purpose it has been used for the separation of globulin from the electrolytes holding it in solution, and the products of the digestion of casein by pancreatin (H. Bechhold*^). The most important recent applications of ultrafiltration are, the separation of colloids with particles of different sizes (fractional ul- trafiltration), and the determination of the colloid or crystalloid nature of doubtful substances. We refer here to the separation of various albumoses by H. Bechhold,*^ the researches concerning the nature of starch solutions by E. Fouard,* and those concerning diastase by Pribram, and the experiment to explain fermentations in the absence of cells by A. von Lebedew,* the researches of Gros- ser on milk (see pp. 174 and 350), the studies of Kirschbaum on dysentery toxin which are still unpublished, as well as those of H. Bechhold on the separation of diphtheria toxin from toxon. By ultrafiltration, Grosser was able to distinguish boiled from unboiled milk (see p. 174). Ultrafiltration is of especial importance in the study of equilibrium in solutions, because in this method there is no change in the balance of crystalloid and colloid portions through the dilution of the solu- tion. It is assumed that only small quantities are filtered, that the differential is in some way ascertained so that no changes in concen- tration occur; and that only moderate pressures are used in the case of solutions containing electrolytes (see p. 59). The numerous researches on iron oxid hydrosol by J. Duclaux and G. Malfitano depend on this, as does the work of R. Burian*^ on salt-albumin mixtures. Ultrafiltration has been variously employed for the solution of purely biological questions. R. Burian*^ has employed it in study- ing the function of the kidney glomeruh, and H. Bechhold*^ in the question of "internal antisepsis." Finally, it must be mentioned, that by ultrafiltration germ-free fluids may be obtained, as well as optically pure water suitable for ultramicroscopic experiments (H. Bechhold*^). New paths have been opened to the study of filterable infectious agents by ultra- filtration (voN Betegh). METHODS OF COLLOID RESEARCH 103 Diffusion. Coefficients of diffusion give information concerning the molecular weight and also the size of the particles of a substance in solution. Diffusion in aqueous solution is the simplest method for such investi- gations. The length of time necessary for such experiments intro- duces so many disturbing factors that, where possible, diffusion in a jelly is to be preferred. A jelly offers a means of separating sub- stances having different rates of diffusion. If a mixture of two sub- stances remain for a time in a tube partly filled with a jelly, the more difficultly diffusible substance will, for the most part, remain in solution and can be poured off, whereas the substance easily diffusible will to a greater extent enter the deeper layers of the jelly. Diffusion experiments in jellies teach us the prop- erties of jellies swollen to various degrees, both in the presence of crystalloids and in their absence. Diffusion in Aqueous Solution. The greatest difficulty lies in avoiding agitation not only when samples are being taken but also during the course of the experiments. The most suitable apparatus is that of L. W. Öholm* (see Fig. 20), with which Herzog made his experiments, and that of Dabrowski. The latter (see Fig. 21a) consists of two glass vessels A and B (a siphon bottle which has been divided in the middle) which are separated by a diaphragm C. This diaphragm is a glass ring filled with glass capillary tubes of 1 mm. bore. The interspaces are filled with celluloid. By this arrangement currents are avoided and a very considerable diffusion surface is obtained. The solution is placed in A and diffuses through C and reaches B from which samples for analysis are taken from time to time through the tube F. The fluid in A as well as in 5 is slowly stirred by the stirrer abd. We shall return to the consideration of Dabrowski' s experiments on the diffusion of albumin with this apparatus on p. 72. In the extremely slow diffusion of colloids, which in the case of the experiments of R. 0. Herzog extended over more than two months, absolute sterility is essential. Besides having sterile vessels, the solutions are also sterilized by saturating them with toluol and layering it over them. The addition of 1/2 per cent sodium fluorid solution is useful also. As previously mentioned, the vessels must Hg Fig. 20. Diffu- sion apparatus. (L. W. Öholm.) 104 COLLOIDS IN BIOLOGY AND MEDICINE stand in a perfectly quiet place; instead of portable water baths, incubators should be used, or if these are not placed so as to be free from vibration, the experiments are kept by preference in a cellar. These diffusion experiments are extremely difficult, but may yield absolutely perfect results, as has been shown by the researches of R. O. Herzog and H. Kasarnowski.* These investigators de- termined the diffusion coefficients for albumin and a number of enzymes (see p. 54 and p. 190), from which it could be determined that they were simple substances. On the other hand it could be shown that clupein sulphate, trypsin and pancreatin were mixtures of different substances. Some of them showed various diffusion layers of dissimilar composition (various percentages of N, in clupein), and with other mixtures, products of different origin showed different coefficients of diffusion (trypsin, pancreatin). Diffusion in a Jelly. A jelly acts like a membrane. It has the advantage over a membrane that its thickness may be varied at will, but the disadvantage that it is usually impossible to obtain the diffused substance in pure form so that the diffused substance must be examined in association with the jelly. Union or adsorption between the jelly and the substance under examination is a more disturbing factor than in the case of a membrane. Diffusion in a jelly has the great advantage over the diffusion in fluids, above de- scribed, that it is not disturbed by currents or the almost unavoidable shaking during the experiment, or while samples are being taken. The experiments are generally performed as follows : A test tube is filled one third to one half full with a very dilute jelly (2 to 5 per cent gelatin). After the jelly solidifies, the solution to be investigated is poured upon it, and the test tube is placed in an ice box. After a longer or shorter time (days, weeks, months) some of the substance will have diffused into the jelly. The supernatant fluid is now poured from the jelly, which is washed with a suitable fluid, water, physio- logical salt solution or the like. The jelly is now examined to deter- mine the course of diffusion, taking the elapsed time into consideration. Gelatin and agar are used as jellies. In many cases, inspection shows the extent to which the fluid has diffused into the jelly {e.g., with dye- stuffs, indicators or precipitation reactions, the results may be seen). For example, gelatin which has been mixed with red blood corpuscles may have tetanolysin layered above it. By the extent of the hemo- lysis it is determined how far the tetanolysin has penetrated. H. Bechhold*^ mixed a jelly with goat-rabbit serum and layered above it a solution of goat serum. The appearance of a white ring showed the distance that the precipitin had penetrated. METHODS OF COLLOID RESEARCH 105 Precautions: For these experiments there should be utiUzed only absolutely pure gelatin or agar which has been dialyzed at least two or, preferably, four or five days in cold running water. Commercial gelatin always contains, besides certain other impurities, sulphurous acid which is used as a bleach in its manufacture, and as a result the gelatin reacts acid to Utmus. It may be completely freed from the acid by neutraUzation with NaOH followed by sufficiently prolonged dialyzation in running water. The dried gelatin is weighed, wrapped in Unen or mull and placed in a trough of running water. After purification the swollen gelatin is very carefully removed from the cloth and weighed to see how much water it has taken up. By the addition of, or the evaporation of water, the jelly is brought to the desired concentration and filtered through a jacketed filter. This is usually a sufficiently accurate method. For absolutely exact investigation, a measured quantity of the moist gelatin must be weighed before and after it has been dried at 105° C. If working with other than pure aqueous solutions, such as with substances which require physiological salt solution to dissolve them, the gelatin or agar must contain the required amount of salt, if, for instance, the diffusion of globuUn solution is desired. Since diffusion experiments with col- loids always extend over a considerable time, the test tubes must be closed with paraffined corks or rubber stoppers. In such cases quantitative measurements may be made with a ruler, by placing the zero at the meniscus of the jelly or by means of cathetometer. Tubes with an engraved scale as arranged by yju^ J J Fig. 21. Diffusion tube. Fig. 21a. Dabrowski's diffusion apparatus. Stoffel-Pringsheim* (see Fig. 21) are convenient. The graduated tube is filled with jelly and the solution is poured into the extensions which are ground on water tight. For accurate measurements the same rules are used as in similar physical measurements. If the diffusion into the jelly is not associated with a visible change, tiie jelly is removed from the glass by placing it in hot water until the periphery melts, so that the cy finder of jelly may be gently pu-shed out of the tube. The jeUy cyfinder is then cut into layers of measured thickness which are studied by chemical, biological or animal experiments as to their content of diffused substance. In this way Sv. Arrhenius* and Th. Madsen have determined the diffusion constants of diphtheria toxin and antitoxin and of tetanoly- sin and antitetanolysin. 106 COLLOIDS IN BIOLOGY AND MEDICINE In order to remove the gelatin cylinder easily, H. Bechhold and J. ZiEGLER coated the interior of the test tube with a lining capsule of parchment, paraffined paper, pergamyn or the like, so that the paper is closed below; on the side, it is closely adherent to the glass, while above it projects about 1 cm. above the rim. This paper lining is filled with gelatin, allowed to cool quickly and removed with the gelatin at the end of the experiment. The gelatin cylinder is sliced after unwrapping the paper. One might imagine that instead of determin- ing the quantity of substance which had diffused into the jelly, i.e., the diffusion path, the percen- tage of substance that has been lost by the re- maining fluid could be determined. For inves- tigations of colloids this method is not to be recommended, because with the slight diffus- ibility of colloids, the loss of substance and the limits of error approach each other closely. The experiments of Voigtländer, placing scales of glue in solutions and determining the amount of the dissolved substance that they took up, are not suitable for use with colloid material. R. Liesegang** has developed a special method. He covers a plate with a jelly and puts on it drops of a solution which diffuse in rings. The method is especially suitable for qualitative studies. If the jelly is impregnated with a substance which forms precipitates with the diffusing solution, structures appear whose form and growth may be beautifully Fig. 22. Osmometer of studied. Instead of aqueous solution, sheets of Biltz and Von Vegesack. ^^^y ^^y ^g placed on the gelatin layer. Diffusion and capillary ascension in filter paper (which must be ab- solutely clean) may, under certain circumstances, give useful quali- tative information. Osmotic Pressure. While in the case of crystalloids, indirect methods of determining the osmotic pressure are used (lowering the freezing point or raising the boiling point), in the case of substances lying at the border line of colloids, the direct osmotic method is most useful. W. Biltz and A. von Vegesack* constructed an osmometer whose main METHODS OF COLLOID RESEARCH 107 feature consists of a collodion membrane (see Fig. 22) protected by a platinum wire netting. In the case of dextrins which lie on the border line between colloids and crystalloids, W. Biltz decreased the permeability of the membrane by adding cupric ferrocyanid. He filled the collodion sac with 1 per cent potassium ferrocyanid solution and placed it in 1 per cent copper sulphate solution. After twenty- four hours he washed the sac for twenty-four hours in running water. The method of procedure recommended by Fouard, impregnation with tannin and gelatin and subsequent tanning with sublimate, has not proven effective, according to Biltz. Above the collodion mem- brane is a glass cap with a vertical tube. The union of netting and cap is at 6. The fluid is mixed by an electromagnetic stirrer c. The electrodes d permit the measurement of the conductivity. The entire instrument is placed in a thermostat. Readings of the rise in the tube are made with a cathetometer. Osmotic Compensation Method This method determines whether crj^stalloids present in a colloidal solution are free or in any way bound, e.g., adsorbed. L. Michaelis and P. RoNA*2 have developed this method in their attempt to solve the question whether the grape sugar, always present in the blood and which, strange to say, does not pass through the kidney, is free or bound in any way. For this purpose we place the fluid to be in- vestigated (in this instance blood) in a fish bladder and suspend it in a glass cylinder containing an isotonic fluid (in this case water with 0.95 per cent NaCl). To the surrounding fluid, in a series of experiments, there is added varying quantities of the crystalloids in question. In this instance sugar is added ; we shall continue to describe the above experiment as an example. If more sugar has been added than is present in the blood it will diffuse into the blood; the sugar content of the surrounding fluid will decrease. If very little or no sugar is added, the sugar will diffuse from the blood into the sur- rounding fluid. This will occur whether the sugars in the blood are free, osmotically active or even if a portion is adsorbed. In the latter case the free sugar will at first diffuse away so that the balance between the adsorbed and the free sugar is disturbed; previously adsorbed sugar may become free and likewise diffuse away. The sugar content of the outer fluid remains constant, only if it accu- rately expresses the free sugar content of the blood. If the total sugar has been determined previously, we may calculate what per- centage is adsorbed or otherwise bound, and what proportion is free or osmotically active. With this method, L. Michaelis and 108 COLLOIDS IN BIOLOGY AND MEDICINE P. RoNA determined that the entire sugar in the blood serum is in free solution. In analogous ways these authors investigated the conditions of union between calcium and the casein of milk. Surface Tension. Because of its sensitiveness, the measurement of surface tension is of the greatest significance for colloid investigation. As far as I know at present, there is no case in which the measurement of these factors has led to the solution of any problem. This is due to the fact that even traces of other substances, especially colloidal sub- stances, markedly influence the surface tension because they are forced to the interfaces. According to J. Traube, HgCl2 in a dilution of one in three million may be detected in dye solutions. For this reason, measurements of surface tension are excessively sensitive and are subject to certain errors. There are two essentially different groups of methods: (a) static, (h) dynamic } (a) Static Methods (the rise of a fluid in a capillary; the def- ormation of an air bubble in a fluid) show the condition of the de- veloped surface, (b) Dynamic methods (the weight or number of falling drops; the pressure necessary to force air through a capillary dipped in a fluid) show the condition of a nascent surface. These two methods, especially with colloids, give fundamentally different values because the interior of a fluid has a very different composition from the surface, and a considerable time always elapses before the surface has assumed its normal properties. As yet, because they are the simplest to perform, only the capillary ascent and the falling drop methods have been used for biological studies. The measurement of the height ascended in filter paper which has been used especially by Goppelsröder in his numerous investigations on alkaloids, dyestuffs and other organic substances ^ may be counted a dynamic, rather than a static method, since in this porous material with the ascent and evaporation, new surfaces are continually formed. Filter paper offers a very useful method for demonstration purposes. Thus, it shows why alcoholic solutions and soap tinctures rather than aqueous solutions are adapted to disinfection of the skin (according to Bechhold) (see p. 404). 1 Detailed descriptions of the methods are to be found in Ostwald-Luther's Physico-Chemischen Messungen (Leipzig, 1910), and in G. Quincke, Poggen- dorff's Annalen d. Physik, 139, 1-89 (1870). 2 KoUoid Zeit., Vol. 4, pp. 41, 94, 191. METHODS OF COLLOID RESEARCH 109 J. Traube has used the falUng drop method with his stalagmom- eter for numerous researches. M. Ascoli has hkewise used it in his meiostagmin reaction in cancer. [Clowes employed it in his studies. See p. 39. Tr.] A given quantity of fluid volume is sucked up into the stalagmom- eter tube and the number of drops required to empty it, dropping it drop by drop, are counted. The stalagmometer is an instrument which requires most careful manipulation to obtain reUable results. It is especially important to keep it scrupulously clean, rinsing frequently with distilled water followed by hot potassium hydrate solution. The apparatus is then placed over night in a hot mixture of concen- trated sulphuric acid and potassium bichromate. Before use it is again thor- oughly rinsed with distilled water. The dropping surface must be absolutely horizontal; this is accomplished by placing it on a stand which can be adjusted in all directions. There must be no bubbles on the dropping surface or in the tubes. Before each initial measurement the fluid to be measured must be sucked up and allowed to flow out again. The number of drops of the fluid is com- pared with the number of the same volume of water. The speed of flow is so gauged that no more than 20 drops fall in a minute. This is best regulated by a screw clamp on a rubber tube slipped over the upper end of the instrument. J. L. R. Morgan has devised a splendid apparatus for determining the weight of falling drops and with it he has measured the surface tension of many substances. There are, however, valuable contributions to the utilization of surface tension (milk investigations by H. Zangger; meiostagmin reaction of M. Ascoli). The separation of colloids and crystalloids hy foaming as well as the separation of colloids of different surface tension is described on page 35. Adsorption. Adsorption experiments may be employed for various purposes. They may be used to determine the distribution of a dissolved col- loid between solvent and adsorbent, thus constituting the determi- nation of a physical constant. In such a case an absolutely chemically indifferent substance, e.g., charcoal, is chosen as absorbent. Ad- sorption presents a suitable means of determining the nature of the electric charge of a dissolved colloid. Positively charged colloids are adsorbed particularly strongly by electronegative suspensions (e.g., kaolin, mastic suspensions); negatively charged colloids are strongly adsorbed by positive suspensions (e.g., iron oxid, clay). Occasionally it is of interest to determine the properties of a gel when it is used as an adsorbent. If in all cases there occurred pure adsorption, whereby a dissolved 110 COLLOIDS IN BIOLOGY AND MEDICINE substance is taken up by a solid one with which it is shaken, the accurate determination of adsorption constants would be of the greatest value. They would then be natural constants of the same class as the boiling points, melting points, etc., which definitely deter- mine the nature of the substances under examination. Unfortunately this is not the case. Chemical phenomena and unexplained factors complicate the pure adsorption phenomena, so that at present, in biological questions, it is only of value to determine whether ad- sorption is the predominating force. Investigations in this field are of great importance. Before the advent of physical chemistry and even now, in biological chemistry, it was usual to search for " pure " substances, and to illustrate a phenomenon by a chemical equation. Adsorption experiments have frequently made it clear to us that in a given case such chemical equations do not and could not exist. The studies of H. Wislicenus on lignin (see p. 248), and on the dyeing process by W. Biltz and H. Freundlich, are selected from among many other classical examples. Adsorption experiments for determining distribution are performed by shaking equal quantities by weight, of the most indifferent solid substance obtainable or a gel (charcoal, cellulose) with various dilu- tions of the dissolved substance. The amount of the substance ad- sorbed is usually ascertained from the solution. It is first determined how much active substance is contained in a unit volume of the solution, which is then examined to see how much has been removed by the adsorbent; the difference gives the quantity adsorbed. Thus H. Wislicenus determined the total solids in the cambial sap of the beech, before and after shaking it with cellulose, and found by taking the difference in weight the amount of colloid that was adsorbed. In individual instances the quantity adsorbed was determined from the adsorbent. B. W. Roux and Yersin treated diphtheria toxin with freshly precipitated calcium phosphate; they then washed the calcium phosphate well and injected it into guinea pigs. A de- termination by means of the adsorbent instead of the fluid I con- sider erroneous in principle, because it has been shown repeatedly, in well-controlled experiments, that the adsorbed substance under- goes changes at the surface of the adsorbent. The fact that a portion of the dissolved substance is removed from the fluid by a solid substance with a large surface does not prove that adsorption has taken place. If, for example, 1 gm. cellulose always removes from a solution the absolutely identical quantity of the dissolved substance, irrespective of the concentration of the solution, we would in all probability be dealing with a chemical phe- nomenon. If the proportion between the adsorbed substance and METHODS OF COLLOID RESEARCH 111 i the amount still in solution remains constant over various dilutions, we may assume that the cellulose forms a solid solution with the sub- stance in question. Adsorption probably exists only if the cellulose takes up almost everything from a very dilute solution and if the absorbing power of the cellulose is markedly decreased with increased concentration of the solution; this condition is frequently observed with dye solutions. Thus we may make shaking experiments with solutions of the concentration, 0.1, 0.2, 0.3, etc., in which 0.1 denotes any arbitrary standard. It must be determined first whether an equilibrium exists at all. For this purpose a given quantity of adsorbent is shaken with the solution, for example, with 100 c.c. In a second experiment an equal quantity of adsorbent is shaken with half the quantity, 50 c.c. of a solution twice the strength. It is then diluted to 100 c.c. and shaken again until an equilibrium is reached. If there is an equilib- rium, the final concentration of the solution in the first case is the same as in the second. If there are material differences, the process may nevertheless be considered an adsorption, but it is complicated by other phenomena as explained on page 27 et seq. If it is unnecessary to determine constants, the simplest proce- dure is to chart the values found on a rectangular system of co- ordinates (millimeter paper). As ordinate is taken the amount of the material that is being investigated which is taken up by 1 gm. of adsorbent (cellulose, charcoal or the like) ; as the abscissa, the amount which remains in solution after the adsorption; so that the curve shows the ratio between the amount of substance in the solu- tion and the amount that is adsorbed. It is easy to determine from the characteristic form of the curve whether an adsorption has oc- curred. (See p. 22.) The determination of adsorption curves and constants is explained in detail on page 22 et seq. It is of greatest importance that the adsorbent be absolutely pure. Many in- vestigators have failed in this and many contradictory results may be attributed to it. The adsorbents are treated with acids, alkaUes, alcohol, ether and benzol according as their nature permits (charcoal, diatomaceous earth or kieselguhr, fibrin, etc.)- In view of the fact that these substances are themselves more or less adsorbed, it is necessary to remove them by prolonged constant treatment with large quantities of the dispersing substance, usually water. Although temperature and time do not play as important a role as in other physico-chemical processes it is important to keep temperature and time con- stant. In most cases the adsorption balance is reached in about one-half hour so that it is always fairly safe to allow an hour. It is usual to shake the adsorbent with the solution, but it must not be over- looked that there are substances which are changed by the mere shaking (see Inactivation by Shaking, p. 34). 112 COLLOIDS IN BIOLOGY AND MEDICINE A second disadvantage of the shaking is that the adsorbent is thereby still further broken up and its surface thus permanently increased. When large quantities of colloid are in solution, there is a counterbalancing error in that the adsorbent becomes coated with a layer of coUoid which thus diminishes the ac- tive surface. Though these errors are small in the case of adsorbed crystalloids, in the case of true colloids they become quite considerable. To eliminate these two disadvantages, H. Wislicentjs and W. Muth* have developed a method which they caU the siphon (or filter) process. In this method a solution of con- stant strength comes repeatedly in contact with the adsorbent. The process is as fol- lows: a tube is filled with washed clay or other adsorbent and in connection with a separatory funnel, forms a siphon. The solution to be studied is poured into the funnel and very slowly filters through the adsorbent. The apparatus (see Fig. 23) is entirely practical. In the strict scientific sense, however, equilibria are not obtained with it. Fig. 23. Apparatus for adsorption analysis. (H. Wislicenus.) Before determining the content of the solution after adsorption, the ad- sorbent must he removed. Filtration is rarely suitable because the filter-paper itself adsorbs. In any event the filter used should be very small, and the quantity of fluid to be filtered as large as possible. Centrifugation is the most practical method. The fluid may be poured or pipetted from the adsorbent which has been deposited. The determination of the content of the solution before and after adsorption varies so much in accordance with the nature of the sub- stance under investigation, that it is hardly possible to formulate general rules. The simplest procedure, when it is possible, is to de- termine the weight of a given volume after evaporation, or the solu- tions may be titrated. In other cases suitable physical or biological methods must be employed (animal experiment, hemolysis, agglu- tination, etc.). To determine the electric charge of a colloid by adsorption, we choose for adsorbent, a suspension of a substance having the most pronounced electrical charge. Electropositive iron oxid or alumina gel removes electronegative colloids from solution. Electronega- tive diatomaceous earth (kieselguhr), kaolin or mastic suspensions (obtained by dropping an alcoholic solution of mastic into water) attract electropositive colloids. As has been said, the charge of METHODS OF COLLOID RESEARCH 113 natural colloids depends chiefly upon their reaction. Experiments are therefore performed with very faintly acid, very faintly alkaline and neutral reactions. Because many substances are destroyed in alkaline or acid solutions, it is necessary to make appropriate pre- liminary tests. Measurements of the amount contained before and after adsorption determine the character of the particular colloid. In this way L. Michaelis investigated a number of ferments (see p. 186). In the border land between adsorption and chemical combination belong the studies of staining, which open to the histologist a wide field for the application of colloid-chemical knowledge. Internal Friction. As has been shown, especially by the investigations of Wolfgang Pauli on albumin, the internal friction or viscosity serves to give valuable information concerning changes in the condition in colloidal solutions. In the case of hydrophile colloids an increase of viscosity usually indicates an hydration. The relative internal friction is usually determined by taking that of water at the same temperature as equal to 1. This is usually done by allowing a given amount of fluid to flow from a capillary tube, taking the time with a stop watch. If the rate of flow for water has been previously determined, the relation between the two gives the relative internal friction. Wilhelm Ostwald constructed a well-adapted apparatus (described in Ostwald-Luther's "Textbook and Manual for the Performance of Physico-chemical Measurements," which see for details). The colloid investigator should not work with capillaries that are too fine, because his fluids are usually very viscous. The maintenance of a constant temperature is of especial importance, and therefore it is necessary to employ a transparent thermostat. Furthermore the specific gravity must also be taken. Ostwald-Sprengel's Pyknom- eter may be used. In biological investigations it is occasionally necessary to work with very small amounts of fiuid. Special apparatus has therefore been devised so that but one or two drops may suffice for a viscosity determination. The apparatus of Hirsch and Beck, thoroughly described by P. T Koranyi and A. v. Richter, " Physical Chemistry and Medicine II," p. 27 et seq., is frequently used. The apparatus of H. A. Determan is very simple; as seen from Fig. 24, it resembles an hour glass. The capillary has at either end an enlargement, and 114 COLLOIDS IN BIOLOGY AND MEDICINE ■Thermome^-er ffffm ^; then a constriction as well as markings. This tube is placed in a large glass shell which can revolve on its axis and has a thermometer inserted. Since the apparatus may be turned upside down like an hour glass, it is possible to take several successive readings from the same quantity of fluid. Determan employs it chiefly for the de- termination of viscosity in uncoagulated blood. For this purpose he places a trace of hirudin on the unbroken skin, preferably on the lobe of the ear. After punc- turing the skin he collects the blood with a pipette directly connected with the tube of the viscosimeter. The apparatus of W, Hess* depends upon a somewhat differ- ent principle. He does not com- pare the time of flow, but the distances fluids may be sucked up. His apparatus consists of two capillaries connected with a T-tube, through which fluids are sucked with a rubber bulb; through one capillary water is sucked, and through the other blood or some other fluid that is to be investigated. From the ratio between the distances to which the two fluids are sucked through the capillaries, the viscosity may be directly determined. The apparatus has certain special advantages; the horizontal position of the capillaries eliminates the influence of the specific gravity; and since water and colloid are simultaneously tested, the errors of temperature are reduced to a minimum and calculations for correction are unnecessary. Fig. 24. Viscosimeter, (H. A. Determan.) Melting, Coagulation and Solidification Temperatures. The determination of the melting,^ coagulation and solidification temperatures has the same significance for colloids as the measure- ment of the melting point has for crystalloids. Coagulation by Heat. The fluid to be investigated is placed in a test tube, in a water bath. The contents of both test tube and 1 In the case of jellies it is only possible to speak of a " period of liquefaction "; for the sake of simplicity I employ the expression "melting point." METHODS OF COLLOID RESEARCH 115 water bath must be stirred and a thermometer must be placed in each. The test tube must be illuminated by a uniform and pro- tected source of light. It is advisable to make a number of prelimi- nary determinations of the coagulation point before making the final reacUng. Wolfgang Pauli distinguishes the following different appear- ances in coagulation: clear, opalescent, slightly cloudy, milky trans- lucent, milky opaque, finely, medium or coarsely flocculent in slightly cloudy or clear fluid. These various aspects are strongly dependent on the dilution and the salt content of the solution, and the latter has the greater influence on the temperature of coagulation. Melting and Solidification Temperature. To determine the melting point of gelatin, agar, etc., W. Pauli and P. Rona* used an apparatus that is similar to that of E. Beckmann for determining the freezing point. The melting point is the temperature at which the layer surrounding the thermometer melts. H. Bechhold and J. Ziegler*- used an air bath, in which a tube containing the jelly is placed alongside the thermometer. The solid jelly is weighted with 5 gm. of mercury.^ The melting point is the temperature at which the mercury breaks through the jelly. Since it is difficult to observe the melting of the jelly and the thermometer at the same time, the authors use an acoustic device (metronome) which is described in the original papers and which is recommended for similar observations. Swelling. The methods of measuring swelling, i.e., the water taken up by a gel, are very inexact. The increase of volume, the gain in weight or the 'pressure of swelling may be determined. Volume Increase. Equal quantities of fibrin may be placed in test tubes and covered with different solutions; we then observe how high the fibrin rises upon swelling (M. H. Fischer, see p. 68, Fig. 7). The increase in volume consists of the decrease in volume of the swelling gel plus the volume of the water, so that the determi- nation has an error, inasmuch as the contraction of the gel during the swelling is unknown. This error is negligible in comparison with the other experimental errors. Increase of Weight. This method introduced by P. Hofmeister is somewhat more accurate. The total solids of the swelling sub- stance (gelatin, muscle, etc.) are determined and the substance either in a dry or a swollen state is placed in a solution. The weight deter- 1 This apparatus is made by C. Gerhard, Bonn, Germany, dealer in chemical utensils. 116 COLLOIDS IN BIOLOGY AND MEDICINE mined before and after the stay in the solution gives the amount of fluid taken up or lost. Before weighing, the jelly is to be wiped with filter paper or a cloth, in order to free it from the adhering fluid. Fluid is pressed out in this method also, especially in the case of very much swollen material, by the weight of the jelly itself as well as by its contraction in the air; this fluid is dried off and cannot be taken account of in the weighing. Swelling Pressure. The deter- mination of this factor offers the greatest prospect for an exact method, especially as the appa- ratus of J. Reinke* may be adapted for other swelling sub- stances. J. Reinke used his apparatus (Fig. 25) to measure the swelling pressure of laminaria (a sea weed). The dry algae are placed in the bore F of the metal cylinder M. On top of the algse rests the piston A, perforated with fine holes. From E water may penetrate to the algae through the holes. As the mass of algae swells it raises the piston and the rod ABD, which may carry various weights. The lifting power is indicated on the dial. The theory of the apparatus, however, requires careful investigation to establish the relationship be- tween sv/elling pressure, water absorbed and the lift. The apparatus of E. Posnjak (see Fig. 26) offers less theoretical difficulties but may be employed only for pressures up to 6 atmospheres. The principle employed follows: the substance to be investigated, Q, is placed at the bottom of a tube G which is closed by a porous clay cell T. The swelling tube dips into a vessel of water. To overcome the swelling pressure the whole vessel is filled with mercury which is connected with a manometer M. The swelling substances may be placed under a given pressure by permitting compressed gas to flow from a steel cylin- der g. The details of the experiments appear in the original paper. Fig. 25. J. Reinkes' apparatus for measuring swelling pressure. METHODS OF COLLOID RESEARCH 117 Flocculation. Observations of the flocculation of a suspension or colloidal solu- tion determine whether it behaves as a hydrophile or a hydrophobe colloid; furthermore, they show the electric charge and, under cer- tain conditions, the presence of a protective colloid. The method of the experiment is very simple: a suspension or colloidal solution is '■6as Cylinder Swelling Substance — Fig. 26. C/ay Thimble divided among a large number of test tubes, diminishing quantities of electrolytes (NaCl, CaCl2, FeCls) are added and the test tubes are filled to equal volumes with a solvent (water, physiological salt solu- tion, etc.). After the test tubes have been exposed at uniform tem- perature for a given time (1 to 24 hours), they are examined for flocculation. In comparative experiments it is necessary that suspensions or solutions have the same concentration. On account of the small amount of sohd substance little is to be accomplished by determinations of total sohds. It is frequently de- sh-able to prepare a large quantity of a standard solution to last a long time, and frequently the determinations may be made colorimetrically. I am accustomed to prepare mastic suspensions by dropping 1 per cent alcoholic mastic solutions into water which is being energetically stirred. The suspension is filtered through rather dense filter paper and tested for transparency in a beaker having parallel sides, to one side of which various printed fines have been glued. The suspen- sion is diluted until, with a definite illumination, a certain size tj^je can just be read. 118 COLLOIDS IN BIOLOGY AND MEDICINE It is necessary to test especially, whether true flocculation has occurred, i.e., whether the particles have really gathered in flocks or whether they have only sunk to the bottom under the influence of gravity. In the case of fine measurements, Jena glass that has been steamed is used, because the leaching out of aUcali from the glass may give rise to errors. This method gives information not only concerning flocculability, but is at the same time quantitative; it informs us concerning "thres- hold values" and, under some circumstances, concerning the rate of flocculation. In order to work with very small quantities, test tubes are used which are narrowed at their lower ends. If there is insufficient fluid even for this determination we may employ a "hanging drop," in other words, drops are mixed on a cover glass, and this is placed up- side down on a slide with a depression ground into it, so that the drop will hang. The cover glass is ringed with vaseline to hold it to the slide. This method is more qualitative and is especially adapted for the agglutination of bacteria. It is desirable in all cases to prepare series of experiments. It may occur as the result of "irregular series" that high and low con- centrations result in flocculation, whereas flocculation may not occur in medium concentration. Electric Migration. Electric migration reveals the nature of the charge of a colloid. The most primitive arrangement for migration experiments is a beaker in which are suspended two platinum electrodes that are part of the circuit of a direct current of at least 60 volts. This is to be recom- mended only for simple demon- strations in the lecture room, where migration must be exhib- ited quickly. Because of the changes in reaction due to elec- trolysis, the results are very in- exact. It is preferable to use a U-shaped tube or an arrange- FiG. 26a. m Simple apparatus for elec- tric migration. ment such as is shown in Fig. 26a. The middle glass jar contains the colloid to be tested, and is united by the U-shaped return bends filled with water to the two outer beakers which also contain water and into which dip the electrodes EE. For research work I employ METHODS OF COLLOID RESEARCH 119 Fig. 27. Bell apparatus. (H. Bechhold.) H. Bechhold's "Bell apparatus."^ The colloid solution to be tested is placed in the glass vessels AA (see Fig. 27) which are connected by a tube. The vessels are closed below by the membranes MM (best for the purpose is fish bladder or the like). The tube R allows for the expansion of the fluid caused by the heat of the electric current. The bell apparatus is placed in two separate glass vessels (crystal- lizing dishes) GG; the mem- branes are immersed in the water, into which also dip the electrodes EE. The advantages of the apparatus are: the great surface of colloid solution; the products transferred do not come in contact with the electrodes and each may be conven- iently and separately collected and ex- amined; the current must pass through the entire colloidal solution; the ap- paratus may be easily sterilized and the free surfaces may have toluol layered over them. The apparatus does not get out of order very easily. L. Michaelis has avoided the change in reaction due to electrolysis at the electrodes by using nonpolarizable electrodes. Fig. 28 amply explains the apparatus. The electrodes, e.g., zinc or silver wire, are dipped into the vessels 1 and 5, which are filled with zinc sulphate and NaCl solution respectively. Migration experiments are usually very difficult to perform. Since the nature of the charge may also be determined by adsorption, by employing positive and negative adsorbents, this latter method is preferable, because it is simpler. Fig. 28. Migration apparatus with non-polarizable elec- trodes. (L. Michaelis.) Optical Methods. There is a certain relation between the cloudiness (Tyndall effect) of a fluid and its content in suspended particles or colloid. On this account various authors (Kamerlingh Onnes and Keesom, Meck- lenburg, WiLKE and Handovsky) have constructed instruments to ^ To be had from the Vereinigten Fabriken für Laboratoriumsbedarf, BerUn N. Schamhorst Str. 120 COLLOIDS IN BIOLOGY AND MEDICINE measure the amount of cloudiness so that they might determine from it, the content of dissolved colloid in the fluid. As yet they have not been applied to biocolloids, and the relation between the clouding of media in a fluid, the intensity of the light and cloudiness yields a complicated curve. On this account it is still impossible to determine the value of these instruments aptly termed by Mecklenburg, tyndallmeters, for the study of biocolloids. There is need of such an instrument. [P. A. KoBER has devised a very satisfac- tory nephelometer which has found ex- tensive application in biology, especially by Bloor. See Journal of Industrial and Fig. 28a. Kober Nephelometer. Engineering Chem., Vol. VII, p. 843. Tr.] I have always felt the want of being able to determine the exact content of a bacterial suspension by some sort of tyndallmeter. Such an instrument must be very simple to manipulate, which is not the case with the existing instruments. The colloid content of a solution is well measured for certain purposes by The Fluid Interferometer. The fluid interferometer^ was originally devised to determine the concentration or change in concentration of crystalloid solutions. According to Marc it is also available for light or yellowish colloidal ^ Made by Carl Zeiss, Jena. METHODS OF COLLOID RESEARCH 121 solutions, but not for those deeply colored. It depends on the follow- ing principle : when parallel beams of light pass through a narrow slit, as the result of refraction a broad band of light with parallel dark bands (interference bands) is seen on the opposite wall. If light is permitted to fall on the same spot through a second parallel slit, the bands of light will interfere and very fine sharp lines will be obtained which may be greatly magnified. When a different medium, that is water or a solution of salt or colloid is placed behind one slit the interference bands move to one side depending on the refractive index. If the process is reversed by a set of glass prisms or something similar, it is possible to read on the adjusting screw of the appa- ratus the difference of refractive index. Fig. 28b shows a cross section through the interferometer. A is the chamber with the standard water, B the chamber for the test solution, C the window for viewing the interference bands. With dilute solutions the concentration increases in proportion to the scale on the graduated drum. For more concen- trated solutions a standard must be set in each case. Technical de- tails of the readings may be found in Marc's paper (he. cit.). He has thus far used the interferometer mainly to determine adsorption and for studying the colloid content in drinking water and sewage. Fig. 28b. Fluid Interferometer. Ultramicroscopy. Ultramicroscopy permits the recognition of certain optical in- homogeneities, and depends upon the use of dark field illumination. Ultramicroscopes magnifying from 750 to 1500 diameters serve in principle the same purpose as the ordinary microscope. They have the advantage over the latter that without staining or extensive preparation, even living objects, spirilla, etc., become visible to the eye; bright on a dark background. Ultramicroscopy with a one hun- dred thousand fold magnification has solved important theoretical questions of colloid chemistry. By reason of the conditions of light refraction its value is chiefly confined to inorganic colloids. In the ordinary microscope the field is usually bright, while the object is more or less dark against its surroundings. In the ultra- microscope, only the rays of light reflected from the object reach the observer's eye and permit the object to stand out bright against the dark background. 122 COLLOIDS IN BIOLOGY AND MEDICINE In this dark field illumination the form of the objects are not given, but every point appears as a small bright disc, which under some circumstances may be surrounded by one or several rings of light. The ultramicroscope is especially suited for the recognition of inhomogeneities in a medium. Apart from the recognition of form, the field of application of the microscope was enormously extended by the invention of the ultrami- croscope. At about seven hundred diameters' magnification, the limit of the available microscopical magnification is reached theoretically and practically, i.e., revelation of new details ceases. The ability to make particles visible in the ultramicroscope is almost unlimited, provided only a sufficiently strong source of light is available. Practically, the limits of visibility in our latitude with the best sunlight is about 10 MM (1 MM = 1 miUionth part of a millimeter). [Zsigmondy gives 5 MM. Tr.j For our purposes, two types must be distinguished: (a) Ultra- microscopes for the study of colloids. They permit the observation of objects or inhomogeneities down to 10 mm and require very bright sources of light — sunlight reflected from a heliostat, or electric arc lights. (&) Ultramicroscopes for the study of organized materials (microorganisms, animal and plant cells), suitable for the study of objects no smaller than 0.1 m- Welsbach or Nernst lights in com- bination with suitable lenses furnish sufficient illumination. Ultramicroscopy for the Study of Colloidal Solutions. The original slit-ultramicroscope constructed by H. Siedentopf and R. Zsigmondy with rectangular arrangement of the optical axes Fig. 31a. Slit-ultramicroscope. is nowadays only employed for the study of solid objects (glasses) and on this account may be omitted from consideration in bio- colloid investigations. Recently Zsigmondy has adapted the original slit-ultramicroscope to immersion. A large proportion of the light METHODS OF COLLOID RESEARCH 123 rays in the path of the object examined are lost by refraction. Very small objects, such as bacteria are too faintly illuminated to be visible by his "dry system." On this account, a highly refractive fluid (water or cedar oil) is placed between the object and the objec- FiG. 29. Illumination of the cardioid ultramicroscope. tive (of wide aperture), which permits many more rays to pass from the object into the objective. It was impossible to use immersion in the earlier sht-ultramicroscope because the illuminating (ßi) and \/^/^^/^^/^/^//^^^^^////////////////A ft * M Fig. 30. Course of the Hght rays through the cardioid condenser. (H. Siedentopf.) Fig. 31. Quartz chamber for the cardioid ultramicroscope. the examining objective (Ss) could not be brought sufficiently close together (see Fig. 31a). This difficulty was overcome through an improved method of construction by the optical works of R. Winkel of GoTTiNGEN (see Fig. 30). A drop of the fluid to be examined is placed between the two immersion objectives of wide aperture or 124 COLLOIDS IN BIOLOGY AND MEDICINE they dip into a small trough containing the fluid for examination. The illumination intensity of the "immersion ultramicroscope " is much greater than the original, and particles are made visible which formerly had quite eluded observation; the contrast effect in the intensely dark field is quite perfect. For biologists, the ultramicro- scope with a cardioid condenser is at present the most important instrument. It permits the use of twenty times as much Hght as the slit-ultramicroscope, "practically the maximum available from the source of light." The construction of the apparatus is shown in Fig, 29. c con- tains an electric arc lamp with a perforated sleeve cap d to cut out Fig. 32. Holder for the quartz chamber em- ployed with the cardioid ultramicroscope. Fig. 33. Flasks for storing ultrawater. (A. Haak.) interfering light. An illuminating lens e passes the light sharply downward, through a glass trough filled with water, to the center of the microscope mirror. The water trough serves to remove the heat rays or when neces- sary acts as a color filter. The microscope mirror throws the light perpendicularly through the cardioid condenser, which replaces the Abbe condenser in the microscope. It is evident from the diagram of the cardioid condenser (Fig. 30) that the various ascending rays strike the slide e obliquely by reason of the double reflection from the two spherical surfaces and that thus, all the light is utilized for illumination; only the rays reflected by the object take the usual path through the objective and the ocular to the observer's eye. Water is used for immersion. With the cardioid ultramicroscope the object is placed between slide and cover glass as in ordinary microscopy. For reasons we shall revert to later, a slide with a special quartz chamber (Fig. 31) is used, which is held in the holder (Fig. 32). METHODS OF COLLOID RESEARCH 125 There are a number of precatitions to be observed in working with the ultra- microscope. Since every impurity makes a point of light in the field, it is neces- sary to employ optically clear water. Such water is prepared according to R. ZsiGMONDY by distillation through a silver condensing tube, or according to H. Bechhold by ultrafiltration through a very tight ultrafilter (6 to 10 per cent). Fig. 34. Dark-field illumination for the examination of organisms. For collecting and storing, only Jena glass vessels should be employed. Ground glass stoppers or corks are to be avoided because they always yield fine dust. I have found the suggestion of W. Biltz serviceable ; he coated the stoppers with Fig. 34a. tin foil. I recommend a storage flask for ultrawater manufactured by A. Haak in Jena (Fig. 33). Neither water nor alcohol should show microscopically the slightest Faraday- Tyndall effect, but wherever illuminated, only a very faint shimmer, whitish in the case of water (ultrawater), bluish in the case of alcohol (ultra-alcohol). 126 COLLOIDS IN BIOLOGY AND MEDICINE Though a skilled ultramicroscopist usually recognizes impurities from irregular intensity of illumination and color of the submicrons, as well as by differences in motion, the most extreme care is necessary in ultramicroscopic work. Cover glasses for the upper chamber should be of quartz, 3/4 mm. thick. The usual methods of cleaning (cloths, brushes, elderpith and Japanese tissue) are to be avoided, as particles, which may cause much trouble, are broken off; scratches, tears, and impurities arising from dry cleaning increase the adsorption of colloids on the chamber walls and reveal their own ultramicroscopic pictures independently of the coUoid particles. The chamber and cover slip must always be cleaned in the following fashion: nothing is touched by hand, only forceps with platinum points or with a loop of platinum wire about them may be used. The apparatus is placed in a hot mixture of concentrated H2SO4 and sodium bi- chromate, then washed with tap water and finally conductivity water. The water must be removed with ultra-alcohol and finally collodion is poured over the cleaned surface. Before use, the collodion skin may be easily raised at one corner and removed. On forcing the chamber and cover slip in the holder, it is necessary to avoid screwing too tight or tensions will arise which gradually equalize themselves and cause striations which are very disturbing. Ultramicroscopes for the Study of Organized Material. The apparatus for this purpose may be adjusted to any microscope. Special preparation of the objects is unnecessary. Objects difficult to make visible by staining or which are too small to see alive with the ordinary microscope are especially suitable for y^ N k II 1 Fig. 35. Abbe condenser with central opacity. Fig. 36. Paraboloid condenser for the dark-field illumination of organisms. (From H. Siedentopf.) this method of investigation, e.g., spirilla, protoplasmic structures, etc. A picture is obtained similar to that with Burri's India ink method, in which the rest of the field is blackened with India ink, while the objects appear bright. Oblique illumination reveals in- homogeneities and structures which would be invisible even with METHODS OF COLLOID RESEARCH 127 staining. In the description of the investigations of N. Gaidukow, E. RÄHLMANN, etc., we shall return to this topic. The optical system depends on the fact that the central rays of light reflected from the mirror of the microscope are cut out by a disc, whereas the lateral rays which strike the object obliquely are utilized. The simplest and cheapest arrangement is the one by which a central blind is placed in the diaphragm carrier of the Abbe illumi- nating apparatus (see Fig. 35), yet this arrangement is not recom- mended on account of the faint illumination and the difficulty in centering. Much to be preferred, because of the strong illumination, is Sieden- topf's paraboloid condenser (see Fig. 36). It is adapted to the study of living bacteria and especially for thin organized structures. The thicker the preparation the weaker must be the objective. Preparations must be made with greater cleanliness than for bright field illumination, though such scrupulous care is not necessary as for the cardioid condenser. The sUde must have a definite thickness (not less than 1.1 mm. or more than 1.4 mm.). The object to be studied is placed on a sUde moistened with a drop of physio- logical salt solution and a cover glass adjusted so that there are no bubbles. The water pressed out at the sides is absorbed and the rim is sealed tight with wax (1 part wax, 2 parts rosin). A drop of water without bubbles is placed between the slide and the condenser. Neither water nor oil is used between the shde and the objective (dry system). [ Other ultramicroscopes have been devised by Cotton and Mouton, and by IvANOWSKi (made by E. Leitz). See also L' Ultra-microscope by Paul Gaston, Paris, 1910. Tr.] An asterisk (*) after an author's name refers to a reference in the index of names. PART II. THE BIOCOLLOIDS. With the exception of water, inorganic salts and a few organic substances as, for instance, urea and sugar, only colloids exist in plant and animal organisms, and if we except water, the colloids quantitatively far exceed the crystalloids. This appears reasonable when we consider the respective roles of crystalloids and colloids in the organism. We may compare living organisms to a city, in which the colloids are the houses and the crystalloids are the people who traverse the streets, disappearing into and emerging from the houses, or who are engaged in demolishing or erecting buildings. The colloids are the stable part of the organism: the crystalloids the mobile part, which penetrating everywhere may bring weal or woe. Because they have only a transitory use, we find in the organism only a small number and a small quantity of organic crystalloids. In plants we encounter the most important organic crystalloid, sugar, on its way from its place of origin to the place where it is used, or in depots, such as buds, roots, fruits, etc., where it is either changed into an insoluble form of carbohydrate, into starches and related products, or its retreat is cut off by the drying of the stem from which the fruit depends. In its course we may tap great quantities of sugar, as in the birch, maple and palm when they are "in sap." If for any reason it becomes mobilized again in the depots, large quantities of sugar may be formed. In wild plants the amount of sugar is rarely very great; it is otherwise with cultivated plants where as the result of cultivation sugar is stored with no advantage to the plant, e.g., sugar beets, sugar cane and common beets. At times a certain biological purpose may be associated with sugar formation, e.g., the sugar formation in fruits for the purpose of their dissemination. The fruit is always the biological object and serves to perpetuate the species, not the individual. The development of a greater quantity of crystalloid as sugar in fruit is therefore not surprising, since the fruit has completed its service for the individual plant. Elsewhere, we find the carbohydrates only in colloidal and most often even in insoluble form. I refer to starches, cellulose and gums. 129 130 COLLOIDS IN BIOLOGY AND MEDICINE Like plants, the animal organism has the power of changing carbohy- drates into crystalloids. Ferments change the starches into sugar, in fact cellulose which is so resistant to chemical attack is made soluble in the intestine of vegetarians, so that it can enter the animal body. As soon as the crystalloid forms of carbohydrate have passed the intestinal wall they are transferred to the main depot, the liver, where they remain in the stable colloidal condition as animal starch, glycogen. We also find glycogen in most of the other organs, where- as the mobile state of carbohydrate, grape sugar, occurs only in minimal quantities (0.08 to 0.12 per cent), in fact only just so much as is necessary for the production of energy. Fats, too, are found in the truly soluble form (e.g., soaps) in plants, only in the germs of seeds; and in animals, probably only at the moment when they pass through the wall of the intestines. They have hardly passed the intestine when they immediately regain their colloidal condition of emulsion, and are carried in that condition to their depots. The same statements hold for proteins. Crystalloid cleavage prod- ucts are found in germinating seeds and in minimal quantities in the vascular paths; in plants, asparagin; in animals, among others, urea, uric acid and ammonia salts. The organism strives its utmost to retain the colloidal condition. Hardly have the crystalloid cleavage products of albumin which have been formed in the stomach and intestines passed through the intestinal wall, than they are straightway changed back into the colloidal form, so that their re- turn may be cut off. The crystalloid combustion products are given an avenue of escape through the kidneys. Physiological chemistry deals with the role of the carbohydrates, fats and proteins apart from water and the inorganic salts. In the study of biocolloids, water and salts cannot be neglected, because water and salts are an indispensable part of the colloids; no colloid can exist in the organism without them, because they condition the turgescence which is characteristic of living colloid. In the case of cells with true membranes, salts may determine at times the balance of osmotic pressure within and without the cell. This general fact does not explain the necessity of the various kinds of anions and cations (K, Na, Ca, Mg, CI, SO4, PO4, CO2); the balance in osmotic pressure may be maintained by any non- electrolyte (e.g., sugar) and yet a cell cannot be kept alive in an iso- tonic sugar solution. Inorganic salts have specific relations to certain organs to which we shall refer later; they are the expression of characteristic sharply defined physical states assumed in the presence of given quantities of water and salt, by the proteins, carbohydrates, etc., of which the organs consist. THE BIOCOLLOIDS 131 .J Chemistry in general, and physiological chemistry in particular, aims to investigate the structure of individual chemical substances, and thus explain their properties by splitting them, synthesizing them, and comparing the regenerated (rearticulated) substance with the original, to see if it is the same or different. Unfortunately, so far as the colloidal constituents of the organism are concerned, they are still far from their goal, especially in the case of carbohydrates and proteins. Here colloid chemistry enters and attempts to com- prehend and where possible to regulate the behavior of the finished product. Colloid chemistry is not occupied with the parts of the machine, but with the machine itself. The chemist splits the pro- teins into Polypeptids, amino-acids, etc., but the student of biocolloids avoids such profound attacks and strives to keep the molecule in- tact so far as possible, studying its outward form, the chemical points of attack offered by the unmutilated molecule, its behavior to changes which may occur under normal and pathological conditions, as well as those brought about by drugs. I wish here to emphasize one other point. Only a few substances occur in the organism that are suitable for study by the physiological chemist. Serum albumin and globulin, the starches and some of the fats, are unquestionably substances which may be separated from the organism without losing some of their essential properties, but they are exceptions. The substances usually studied by physiological chemists are those which have already suffered considerable modi- fication. The organism possesses neither glue, histone nor myosin, and even if we knew the exact chemical constitution of glue, this would throw no light upon the properties and the function of cartilage and the fibrils of connective tissue from which it is derived. But even without knowing the chemical composition of glue, I believe that it would be possible, with the methods of colloid chemistry alone, to collect a series of observations which would afford valuable con- clusions concerning the chemical mechanisms of such tissues. A time will come when the old physiological chemistry and the new chemistry of the biocolloids will meet and the two opposite ends of the tunnel shall be united. We shall first try to learn the properties of the intact colloid molecule of the colloid particle. The following chapters on carbohydrates, lipoids and proteins should be read, bearing this statement in mind. CHAPTER VIII. CARBOHYDRATES. As the name indicates, we classify as carbohydrates a group of substances containing carbon and the elements of water, i.e., 0 and H in the proportion of 1 : 2. We owe our knowledge of the constitution of the lower members of this group, the crystalloid water-soluble sugars, largely to the in- vestigations of Emil Fischer. The same difficulties which we encounter in the study of all colloidal substances interfere with de- termining the constitution of the higher colloidal members of this group, the saccharocolloids. There is at present no means of positively recognizing the purity and the individuality of the substance studied or its derivatives. It is true '^that we may crystallize individual colloidal carbohydrates, e.g., inulin, which as a rule naturally occurs in crystals, but all we have said on page 71 concerning the crystalli- zation of colloids in general applies to inulin. Because of their common occurrence, the most important sac- charocolloids are the starches, vegetable and animal (glycogen), and also cellulose. Next in importance come the various gums and pec- tinous plant juices. Dextrins which are also usually colloidal are really cleavage products of the starches. A host of individual facts have been derived from the enormously extensive utilization of starches, as food, cereals and potatoes, for fermented liquors, beer and brandy, as sizing, etc., and of cellulose (in the textile industries and paper manufacture). It is only recently that there has been manifested an effort to reach a general view- point such as colloid science has made possible. (E. Fouard.*) Starch, obtained from starchy grains, is an amorphous white powder which migrates in the electric current to the anode, it exhibits an acid character chemically, since it adsorbs dissolved alkalis (with the exception of NH4OH) and hydroxids of the heavy metals, probably thus forming amylates. It does not adsorb acids or salts. (A. Rakowski.*) Since phosphoric acid is always present in native starches and in the diastatic cleavage of phosphorus-containing dextrins, we may assume with M. Samec that there is a carbohydrate phosphoric acid complex probably an ester (amylophosphate) . Starch has a great reversible swelling capacity in water (pore swelling). 133 134 COLLOIDS IN BIOLOGY AND MEDICINE In swelling there is a great loss of volume, i.e., the volume of the swollen starch is less than that of the dry starch plus the water necessary for swelling, as was shown in exhaustive experiments by H. RoDEWALD.* This contraction is about 8 per cent, when 20 per cent of water is taken up. Swelling is accompanied by the liberation of heat, which, according to E. Wiedemann and Charles Lude- KiNG,* amounts to about 6.6 calories per gram. H. Rodewald studied the phenomenon more thoroughly and found a diminution in the amount of heat liberated with increasing water content. The following is an abbreviated table of his results: Per cent of watet contained in 100 gm. dry starch. Approximate per cent of heat Hberated per gm. dry starch. 0.23 28.11 2.39 22.60 4.58 18.19 9.59 10.28 18.43 3.54 If we add more water to starch, and heat to 55°-70° C, by "solu- tion swelling," we get a jelly-like mass, starch paste, which dissolves on continued heating in more water. This solution coagulates when it is frozen. G. Malfitano and A. N. Moschkoff* utilize this property of starch solution to obtain a starch free from mineral substances. Demineralized starch on being mixed with suitable salts shows all the properties of the different forms of starch. These investigators are therefore of the opinion that the various modifica- tions in the properties of the natural starch granules are due to mineral admixtures. E. FouARD,* by means of acids, freed starches from their inor- ganic elements and obtained a substance which formed an unstable colloidal solution in water. Heat, alkalis and alkaline salts made the solution more permanent, whereas cold, acids and acid salts favored jelly formation. On ultrafiltering his starch solutions, E. Fouard found that in accordance with their concentration, a given fraction of the solution always passed through collodion membranes. He concluded from this, that for every concentration of the starch solu- tion a balance exists between the coarser particles and the molec- ularly dissolved (hydrolyzed?) starches. Unfortunately, the work of E. Fouard contains no information relative to the permeability of the collodion membranes, so that it is impossible to arrive at any conclusion in reference to the size of the suspended and the dissolved starch particles. CARBOHYDRATES 135 On account of their great surface development, the adsorptive capacity of starches is very great. As has been said, when they swell they adsorb water, dyes, etc. A very characteristic adsorption compound is formed with iodin. lodin is the best known reagent for starches; by it they are stained blue. It was formerly beheved that iodin and starch united chemically; W. Biltz showed that it is merely an adsorption. According to the degree of dispersion, iodin solution is blue, red, orange or yellow, inasmuch as the starch solu- tion acts as a protective colloid (W. Harrison*). There are, in addition, varieties of starch which give at once a brownish red or a wine red color with iodin. Inulin and lichenin are colored yellow by iodin. The swelling and pasting of starches, hydration, is analogous to the swelling of proteins, which is a preliminary to their hydrolytic cleavage. The swelling of starches is favored by electrolytes, especially alkalis, so that swelling commences at a much lower temperature in their presence. For this purpose the anions are especially important and in fact, in a lyotropic series, similar to that of acid albumin. See page 152 (M, Samec).* Starch paste increases the surface tension of water (Zlobicki*). A solution of starch in water, as well as one of dextrin, dissolves less CO2 than pure water (according to A. Findlay*). (A gelatin solu- tion dissolves more CO2 than pure water!) 'Under the influence of dilute acids or diastatic ferments, the starch molecule takes up water and, step by step, breaks into small frag- ments, soluble starches, amylodextrin, various dextrins some of which crystallize, and finally into grape sugar. The larger the fragments the more marked is their colloidal character. As the result of osmometric experiments W. Biltz * arrived at the following molecular weights: Amylodextrin 22,200-20,500 Higher achroodextrin 11,700- 8,200 Erythrodextrin 6,800- 3,000 Acid dextrin 4,000 Lower achroodextrin 1,800- 1,200 Dextrin (C6Hio06)6 905 Commercial dextrin 6,200- 2,700 "Soluble starch" (according to H. Friedenthal*^ produces a definite lowering of the freezing point, which is proportionate to the amount of the substance that is dissolved. CrystalUzable dextrins [amyloses (C6Hio05)6] prepared by H Pringsheim* and Eissler combine with iodin to form iodin-addition compounds which dissolve like iodin starches in cold water with a transitory blue color. 136 COLLOIDS IN BIOLOGY AND MEDICINE Commercial dextrins which are mixtures of starch fragments of different size are almost entirely held back by impermeable ultra- filters (10 per cent) (H. Bechhold*^). Closely related to the starches is inulin, the reserve carbohydrate in dahha bulbs and the roots of Inula helenium, etc., as well as lichenin, which occurs in many lichens, especially Iceland moss. Unlike the starches, inulin and lichenin are soluble in water without forming a paste and form yellow adsorption compounds with iodin (see p. 135). Besides these, a series of starches have been identified, some of which show differences in their final cleavage products, the sugars. As yet they have not been studied colloid-chemically. In its biological function, animal starch, glycogen, resembles the plant starches closely, and in its colloid properties stands midway between these and inulin. It swells in cold water and forms with it an opalescent hydrosol. The electric current carries it to the anode (Z. Gatin-Gruszewska*). With iodin it forms according to its concentration, a brownish yellow to deep red adsorption com- pound. The internal friction of glycogen solutions have been studied by F. BoTTAZzi and G. D'Errico* as well as by J. Friedländer.* Glycogen is split up by acids and ferments, and according to the degree of hydrolysis we find all sorts of fragments, from the highly colloidal to the easily diffusible grape sugar. E. Rählmann*^ fol- lowed this process with the ultramicroscope. The glucosides must be mentioned in this connection. They are compounds of the aliphatic and the aromatic series with sugars, which may be split into their components by acids or ferments. In the vegetable kingdom they include very active pharmacologic and toxic substances, such as digitalis glucoside, phloridzin and saponins. Recently several glucosides have been discovered in the animal organ- ism, e.g., cerebron in the human brain. Though some glucosides, e.g., amygdalin and myronic acid are unquestionably crystalloids, others, e.g., saponin, are entirely colloidal. Since we know very little of the biological significance of glucosides, it is evident that we do not know what importance may be ascribed to the crystalloid form in one and the colloidal form in the other. The gums are carbohydrates which are widely distributed through- out the vegetable kingdom. Some of them play a part, in many respects analogous to that of fibrin in the animal kingdom, since they solidify on issuing from a wound, thus sealing it. Best known of the gums are gum arabic, carraghen and cherry gum, while agar, de- rived from Japanese sea weed, is of especial importance in bacteri- CARBOHYDRATES 137 ology. Finally, we must mention the pectinous plan-t juices, which unlike the true gums are slightly or not at all soluble in water. The gums are typical examples of hydrophile colloids; they swell into jellies in water, and on adding more water pass, at an indefinite point, into solution. Rise of temperature shifts this point in favor of solution, though it is by no means immaterial at what condition of swelling the heating occurs. If, for instance, agar has been allowed to swell in cold water for a long time, it immediately becomes a homo- geneous solution on warming. If solid agar is heated in water, we get a lumpy suspension of agar in water, which only very gradually becomes a homogeneous sol. It is evidently necessary for each particle of agar to have the amount of water necessary for solution in close proximity before it is warmed; otherwise the swelling will occur but slowly from the outside, where there is an excess of water, and proceed inward, since the peripheral particles of agar hold the water until they are dissolved. Indeed, the phenomenon is one which depends on the size of the surface; the large mass with relatively small surface dissolves more slowly than the same mass divided, i.e., with a relatively increased surface. Solutions of gum do not dialyze. In my opinion little attention need be paid to the determi- nation of their osmotic pressure, since traces of electrolytes which cannot be removed, suffice to simulate it. I know of no studies on the electrical migration or on the diffusion coefläcients of gums. [W. M. Bayliss has recently determined the viscosity and osmotic pressure against water and Ringers' solution of gum acacia, gelatin and amylopectin. He recommends the use of gum and gelatin in saline infusions as a method of maintaining blood pressure. The more prolonged action of such infusions he attributes to the osmotic pressure of the colloids. Proceedings of the Royal Society of London, Series B, No. 89, pp. 380-393. Tr.] Gums usually diminish the surface tension of water. The tr of a 20 per cent solution of gum arable is 9 per cent lower, and a dilute solution of agar 5 per cent lower than that of water (G. Quincke). Some kinds of gum increase the surface tension of water (Zlobicki*). The general facts, stated on page 66, hold for the swelling and shrinking of gums. On sweUing, the heat hberated, according to E. Wiedemann and Chas. Ludeking,* is 9.0 cal. per gm. for gum arable and 10.3 cal. per gm. for gum tragacanth. Wo. Pauli*^ found that a rise of temperature accompanied the sweUing of carraghen. The significance of crystalloids for swelling and turgor has been studied chiefly in gelatin. In the case of the gums, other than agar, no investigations of this point have been made. Though the prob- ability of many similarities exists, an absolute parallelism cannot be 138 COLLOIDS IN BIOLOGY AND MEDICINE assumed. Thus, for instance, the melting point of gelatin is raised by grape sugar and glycerin, whereas that of agar is reduced. NaCl elevates the melting point of agar and depresses that of gelatin (H. Bechhold and J. Ziegler*^). Agar has a very strong tendency to gelatinize; even 1 gm. per liter gelatinizes at 0°. This great gelatinizing capacity led Robert Koch to make his culture media of agar, and permitted him to grow cultures of bacteria on solid media at body temperature. Gelatin media which had been used at first melt at 37° C, and could, ac- cordingly, only be used at room temperature. Electrolytes as well as nonelectrolytes alter the gelatinization time of agar. Nitrates, iodids, sulphocyanids, benzoates, urea and thiourea lengthen it; chlorids, bromids, acetates and salts of poly- basic acids shorten it. Cellulose is for plants what bones are for animals. It forms the framework which maintains their shape. If it is to fulfill this function it must be insensitive to the chemical influences of the plant juices, and must not be able to swell. Wooden relics are by no means un- common; only in exceptional instances are fats and proteins or gelatinous constituents seen after thousands of years, and then only under very unusually favorable conditions, as in the desert climate of Egypt. Wood, even uncarbonized, is a common relic not only for Egyptian archaeologists and travellers in the Turanian deserts, but it has frequently been preserved in our own climate and even in water. Oak bridge piles dating from Roman times have been found in the Rhine, wood carvings and wooden buckets in the springs of Salzburg, fragments of boats of the lake dwellers, in the Swiss lakes, and those of the Vikings in the peat bogs of North Germany and Jutland. Stability of form, in other words, a slight swelling capacity, makes wood, next to stone, metal and bone, suitable for many pur- poses. Cellulose, the principal constituent of wood, is extremely inactive and is only split up into soluble sugars (chiefly grape sugar) by strong chemical action (acids concentrated or under pressure), or by specific ferments (bacteria in the intestines of ruminants). Cellulose not only has a high adsorptive capacity for dyestuffs, but even true suspensions are fixed at its surface. For this reason cellulose has recently been used like charcoal a's a clarifier and as a filter for turbid liquids. CHAPTER IX. LIPOIDS. '' Lipoids " is the collective name for fatty substances.^ Many of them are not moistened by water; however, this property is not characteristic of all lipoids. Fats and oils are esters of higher fatty acids, usually with glycerin, which may be substituted by other higher alcohols; for instance, a palmitic acid ester of cetyl alcohol occurs in spermaceti, found in the skull of the sperm-whale. Though in other fats all three hydroxyls of glycerin are replaced by fatty acid radicals, in the lecithins only two fatty acid radicals occur, and the third hydroxyl group is re- placed by a phosphoric acid-cholin radical. Cholin is a trimethyl- oxy ethylammoniumhydroxid . Formula of Fats. Formula of Lecithin. CHzO-fatty acid CHaO-cholin-phosphate CHO-fatty acid CHO-fatty acid CHsO-fatty acid CHaO-fatty acid Finally, we must consider Cholesterins and isocholesterins, which we may regard as complex terpenes. The characteristic fats, the triglycerides, are universally distributed in the animal body, where they play an important part in maintaining the body heat, while in plants they are of much less importance. Lecithins are found distributed throughout the animal organism, not only in the chief depots, the brain, nervojiis tissue generally and the egg yolk, but in every cell, every organ, even in the lymph, blood corpuscles and muscles. In plants too, lecithin is widely distributed, occurring in the seeds. The fact that lecithins occur in all parts of the body is an evidence of their great biological importance. So far as may be gathered from previous researches, they play an important role in the life- 1 Various investigators give different definitions of the term " lipoids." Bang uses it most inclusively and regards everything in the body soluble in organic eolvents as lipoid; S. Loewe gives it the narrowest scope, and includes only sub- stances which form colloid solutions in organic solvents {e.g., cephalin, cerebrosid). 139 140 COLLOIDS IN BIOLOGY AND MEDICINE processes of the cells and in the adjustment of the metabolism be- tween cells and their surrounding media. The same is true of Cholesterin which is frequently associated with them. Fats and oils are not soluble in water and aqueous solutions; but instead they are easily emulsified by a great variety of substances. A few drops of lye suffice to make the finest sort of subdivision of oil in water. It is still an open question, whether this is accomplished by the lye itself, or whether it is due primarily to soaps, which are formed from the free fatty acids always present in fats and oils, and which themselves act as emulsifiers. Soluble soaps, i.e., the fatty acid salts of the alkalis, possess remarkable fat-emulsifying proper- ties; this property is also shared by the intestinal juice, the pan- creatic juice and the bile. Emulsions of fat and oil usually occur in alkahne solution, while on the other hand acids produce floccula- tion. There are exceptions to this, e.g., the lipase of the castor bean emulsifies fat in acid solution, and milk curded by rennet yields a stable acid emulsion on digestion in pepsin-hydrochloric acid. In general, fat emulsions behave like hydrophile colloids; they are not as easily coagulated by neutral salts as are hydrophobe colloids or other suspensions. Milk is a natural emulsion of fat (see p. 345 et seq.). Though in the examples given so far, fat has been the dispersed phase and water or the aqueous solution the dispersing medium, conversely, water and aqueous solutions may be incorporated in fats. In this case fat is the dispersing medium and the aqueous solution the dispersed phase. Instances of this condition are butter, cold cream, which is cooling because of the water it contains, lanolin, as well as many salves and liniments. Structures like cream and whipped cream occupy a characteristic intermediate position. Lecithin behaves in a very peculiar way. It forms an emulsion with water of its own accord; indeed like a protein it swells up in water into a turbid colloidal solution, without dissolving. It may be said that it occupies a place, in respect to its colloidal properties, between the emulsifiable fats and the hydrophile colloids, closely ap- proaching the latter. 0. Forges and E. Neubauer* studied its properties by experi- menting upon the coagulation of lecithin emulsions. The precipitating action of neutral salts is in a lyotropic series similar to that for acid albumin, in which the greatest effect is pro- duced by the anions. Salts of the alkaline earths and the heavy metals frequently yield "zones of inhibition" as described on page 84. It is remarkable that neither HgCU nor Hg(CN)2 even in -p LIPOIDS 141 jf Concentration cause precipitation. This is in thorough accord with the solubility of such substances in fats. Lecithin acts towards colloids and suspensions (ferric hydroxid, mastic suspension) like any other colloid which migrates to the anode. Similarly charged colloids cause no precipitation (and lecithin may even act as a protective colloid for mastic) ; oppositely charged colloids produce flocculation in suitable mixtures (ferric- oxid hydrosol). Saponin clears lecithin suspensions. Alcoholic lecithin solutions are much more stable in the presence of salts than aqueous solutions. Mercuric chlorid is an exception. Alcoholic lecithin solutions protect some other colloids, e.g., albumoses, from the precipitating action of alcohol. (L. Michaelis and P. IlONA*^.) Ethereal lecithin solutions cause some otherwise insoluble sub- stance to dissolve in ether (e.g., NaCl and grape sugar). This prop- erty is evidently due to the fact that in ethereal solution, lecithin has a great capacity for taking up water. Cholesterin, according to the investigations of 0. Porges and E. Neubauer,* is a hydrophobe colloid. Its aqueous emulsion behaves like a mastic suspension in the presence of a large variety of salts. The same is true for its behavior with other colloids. In neutral solution it is precipitated by certain proportions of albumin and saponin. Lecithin may act as a protective colloid for Cholesterin. Cholesterin forms a true solution in alcohol and ether, and in such solutions exhibits no colloid precipitation reactions. CHAPTER X. PROTEINS. We designate as proteins a group of nitrogenous colloids which are the chief constituents of animals and plants. They consist en- tirely or chiefly of substances which contain quantitatively: Per cent. C 50-55 H 6.5-7.3 N 15-17.6 0 19-24 S 0.3- 2.4 One of the chief characteristics of most of the dissolved albumins is their coagulability when heated. The effect of heat on undis- solved proteins is shown by the loss of their capacity to swell; they are "denatured." Hydrophile colloids become hydrophobe. A host of the most diverse substances are included under the generic term "albumin." It includes water-soluble substances such as egg and serum albumin, and substances soluble in saline solutions, as globulin, vitellin, myosin and, finally, such substances as are solu- ble neither in aqueous nor in saline solution, for example, fibrin. We know that there exists in each plant and in each animal a distinct serum albumin and a distinct serum globulin, etc. In the chapter on "Immunity Reactions," we shall return to the species-native charac- teristics (Artspezifität) of proteins (see p. 194). We shall not speak of these distinctions here, but we shall dwell, rather, upon the prop- erties that the different proteins possess in common. Colloid research, in a negative way, by destroying a large number of false conceptions, has been of great service to the chemistry of proteins; and it is in a position to establish new principles, since only a few proteins crystallize and, with others, common methods of purification are unavailable. Absolutely misleading methods have been relied upon to separate and distinguish proteins. It was form- erly believed, e.g., that the coagulation temperature of different proteins varied, but colloid investigations demonstrated that small quantities of electrolytes could raise or depress it to a great extent. By pre- cipitation with copper sulphate, E. Harnack believed that he had obtained characteristic copper albuminates, and other observers that they had obtained characteristic silver or calcium albuminates. Colloid chemistry has shown that the different amounts of copper, 142 PROTEINS 143 silver, etc., contained in such precipitates depend upon the concen- trations of the solutions of albumin and of electrolyte, and that precipitates of constant constitution are always obtained under the same conditions. Fr. N. Schulz and R. Zsigmondy showed that crystallized egg albumin which had adsorbed colloidal metallic gold, recrystallized with it. As the result of such observations we become very sceptical con- cerning the ''purity" of proteins. However, it is just such explana- tion of earUer errors which shows us upon what facts we may really depend, and gives to science a new method and, in part, a new course. Before we describe the few proteins which have been studied colloid- chemically, we shall consider briefly some of their general properties. One of the most characteristic properties of many proteins is coagulation. It may be brought about, either by a rise of temper- ature (heat coagulation) or by chemical means. Most of the coagulations due to the salts of the light metals and some of those due to the alkaline earths are reversible, i.e., the coagulations reverse themselves by the addition of more water. Heat coagulation and coagulation due to many of the salts of the heavy metals are irreversible. The coagulations due to alcohol, acetone and ether are intermediate, that is, the coagulation produced is at first soluble in water but becomes insoluble after a while. Globulin which has been preserved for a time in pure water behaves in a similar way, for it then becomes less soluble in salt solutions. Though reversible coagulation may be viewed as a purely physical salting out (see under this heading) a chemical change must be assumed in the cases of irreversible coagulation. Many heavy metals form insoluble complexes with albumin (see p. 157).^ Irreversible coagu- lation by heat, alcohol, etc., may be explained, possibly, by a chemical transformation. The fact that the H ion concentration diminishes after heat coagulation is in favor of this view (Sörensen and Jurg- SEN,* H. Chick and C. J. Martin,* Guaglieriello*) . In the case of heat coagulation, water appears to enter the albumin molecule, because absolutely dry hemoglobin and egg albumin may be heated to 120° C. without losing their solubiHty in water (H. Chick and C. J. Martin*). Possibly this is the initial stage of hydrolysis, since according to Berczeller* the surface tension of salt-poor albumin solutions is temporarily depressed upon boiUng, just as occurs upon hydrolysis by pepsin, trypsin, etc. Irreversibly coagulated albumi- nous pellicles may be formed merely by shaking with air (see p. 34). 1 [Sansum has shown that after the absorptionof alethaldoseof 4mg. perkilo of mercuric chlorid, no treatment avails. Jour. Am. Med. Association, vol. 70, p. 824. Tr.] 144 COLLOIDS IN BIOLOGY, AND MEDICINE Though native albumins are usually hydrophile, they become hydro- phobe upon heat coagulation. Trace's of acids and salts cause precipita- tion. The precipitate of albumin induced by freezing is irreversible. Albumin may be partly changed to globuhns, and ultimately coagulated and precipitated by light, particularly light of short wave length (G. Dreyer and Hausen, Chalupecky), ultraviolet rays are particularly intense in their action (Bovie). This is especially significant for some future explanation of the action of sunlight on the organism. Schanz attributes to it the clouding of the crystalline lens in cataract. The rays of shortest wave length, the Roentgen rays, coagulate albumin. A number of proteins have been crystallized (e.g., egg albumin, horse serum albumin, hemoglobin, aleuron) and though the shape of the crystal is characteristic for the kind of albumin, nevertheless it is impossible to obtain the crystals absolutely chemically pure as in the case of crystalloids (see p. 71). Albumin solutions have been studied ultramicroscopically by E. Rählmann,*^ E. von Behring, H. Much, Römer and C. Siebert,* by L. Michaelis,*^ L. Pinkussohn* and J. Lemanissier.* The results expected at the outset were not realized, so that, in recent years, there has been little heard on the subject. In my opinion this is unfortunate; I am inclined to believe that valuable data might be gleaned from a properly controlled ultramicroscopic study of proteins. It is evident that a large part of albumin solutions is amicroscopic, so that only such portions are seen as show a different refraction than water or physiological salt solution. An albuminous solution shows a different number of ultramicrons, entirely depend- ing upon whether it has been prepared in water or in physiological salt solution (Michaelis); and with different dilutions depending upon the salt content, a different number of small particles become visible (Rählmann). On this account L. Michaelis and J. Lema- nissier do not share the opinion of E. Rählmann and the school of E. VON Behring as to the suitability of ultramicroscopic observa- tion for the quantitative determination of albumin, e.g., in the urine. Great interest must attach to ultramicroscopic observations of the cleavage of albumin by pepsin,^ the influence of therapeutically active substances (ferric chlorid, alum, tannic acid, silver nitrate, copper sulphate, collargol, etc.), as well as the effect of dyes on solu- tion of albumin (Rählmann). A few submicrons were found by J. Lemanissier in albumin solution and many in hemoglobin, but they disappeared in 24 hours. Ultrafiltration of albumin solution is still in its infancy. H. Bechhold has shown that the particles of serum albumin are some- 1 [Already observed by J. Alexander. Jour. Am. Chem. Soc, Vol. XXXII, p. 680, et seq. Tr.] PROTEINS 145 what smaller than those of hemoglobin. Unlike ferments, proteins are not strongly adsorbed by filter material. All albumins are amphoteric electrolytes, i.e., they yield H and OH ions; otherwise expressed, they have at the same time the character of weak acids and of weak bases, with the acid character more or less in excess. The consequences resulting in the case of albumin have been discussed more extensively on p. 154. The isoelectric point is that where the sum of the H and OH ions is least. This point acquired especial significance from the studies of L. Michaelis who showed that the isoelectric point was charac- teristic for each albumin. That albumins are most easily precipitated at this point was also demonstrated [by Hardy. Tr.]. In this respect they behave like crystalloid electrolytes. Neutral molecules are much more difficult to dissolve than their ions. Acids sUghtly dis- sociated electrically, e.g., uric acid, salicylic acid, quinine, are much more difficult to dissolve than their strongly dissociated salts. Adsorption phenomena are of great importance. Proteins may be strongly adsorbed or, on the other hand, exert a powerful adsorption. The purely physical phenomena are complicated by the intermingling of specific chemical properties and thus very decided differences be- tween the various groups of albumins are brought to light. Proteins as Adsorbed Substances. Adsorption has been most carefully studied in the case of albumin. As a result of its faint acidity it is completely adsorbed by ferric oxid hydrogel, but mastic and kaolin suspensions on the contrary adsorb it only in faintly acid solution (L. Michaelis and P. Rona*). On this account, any suspension may be employed to remove albumin from acid solutions, e.g., urine, whereas an electropositive adsorbent {e.g., ferric oxid gel) must be chosen in the case of neutral fluids. Although the distri- bution between solvent and adsorbent has the shape of an adsorp- tion curve, it must nevertheless be emphasized that the process (adsorption by iron-oxid, cellulose and kaolin) is only incompletely reversible, thus resembling the phenomena of dyeing (W. Biltz*^). The adsorption of euglobulin by kaoHn (K. Landsteiner and Uhlirz*) is to be explained in a similar way. Proteins as Adsorbents. Proteins are frequently used as adsorb- ent both in a solid and in a denatured condition. They take up acids, alkahs, salts, dyes, etc., from solution, in accordance with the formula of an adsorption curve. In my opinion it is best to regard the compound as an adsorption whenever the chemical constitution of the adsorbed substance is unknown or when it, itself, possesses col- loid properties. To view the facts from the standpoint of chemical constitution (see p. 154), a viewpoint which presupposes a more 146 COLLOIDS IN BIOLOGY AND MEDICINE exact knowledge of the mechanism of the reaction, seems to me to be a still more advanced step. Adsorption by protein in solution is more important than ad- sorption by solid proteins. By ultrafiltration it might be possible to investigate the distribution between a dissolved colloid and a crystal- loid. In this connection I am acquainted only with the investiga- tions of H. Bechhold on the distribution of methylene blue between water and serum albumin (see p. 26). Thomas Geaham and R. O. Herzog*^ determined the coefficient of D cm^. diffusion of egg albumin and ovomucoid to be f- • 10^. Its "^ seconds values are in the case of Egg albumin 0.063 (at 13° C.) measured by Graham, calculated by Stefan. Egg albumin 0.054 (at 15.3° C.) according to Herzog. Egg albumin 0.046 (at 7.75° C.) according to Herzog. Egg albumin [crystallized with 3.6%. .. .0.081 (at 16° C.) according to Dabrowski* (NH4)2S04]. Ovomucoid 0.034 (at 7.75° C.) according to Herzog. Glucose (for comparison) . 0.57 (at 18° C.) From these figures the radius r of albumin particles has been calculated for Salt-free egg albumin 2,43 mm Crystallized egg albumin 1,37 ßß [with 3.6% (NH4)2S04] This diminution in the size of the albumin particles in the presence of (NH4)2S04 coincides with what we shall learn of the other effects of neutral salts on albumin (see p. 151). When solid albumins go into solution there occurs a diminution in volume amounting to about 5-8 per cent, as is the case with starches. (H, Chick and C. J. Martin.*) Egg albumin and serum albumin, globulin, casein and fibrin have been most carefully studied colloid chemically. ALBUMINS. Albumins are soluble in water, and in dilute neutral salt, and in acid and in alkaline solutions. They are usually found in the com- pany of globulins and there are reasons for beheving that they may be converted into globulins by moderate heating. Albumins occur almost exclusively in serum, in eggs and in milk; the existence of plant albumins is not yet definitely established. PROTEINS 147 In the organism proteins only occur accompanied by electrolytes which greatly modify their properties. On this account we shall try to get an idea of albumin unassociated with electrolytes in order to un- derstand the influence of the addition of electrolytes. Electrolyte-Free Albumin.^ Wolfgang Pauli obtained an albumin free from electrolytes by dialysing ox serum in closed vessels for eight weeks. After standing undisturbed for several weeks, the serum was filtered and was foimd to furnish a stable crystal-clear fluid. Boiling and the addition of alcohol completely coagulated the solution. Such albumin is am- photeric with a weakly electronegative charge; so that it consists chiefly of neutral and very slightly ionized particles ^ which migrate to both electrodes in an electric field (L. Michaelis*^). According to L. Michaelis and P. Rona, the isoelectric point for serum al- bumin, at which there is the greatest tendency to precipitation, occurs with an H-ion concentration of 2.10~^; for boiled, denatured serum albumin when the H-ion concentration is 4.10~^. It increases the internal friction of water considerably. If the friction coefficient of water is represented by 1000, a 1 per cent amphoteric albumin solution at the same temperature will be 1068. An equimolecular 1 per cent salt solution causes no demonstrable change in the coeffi- cient of friction of water. Solubility in Albumin Sols. We shall see in the following pages that albumin usually has a powerful influence on the solubility of substances. It is a remark- able fact that the solubility of carbonic acid is the same in an albumin sol as it is in water (A. Findlay*). This is all the more re- markable since starches and gelatins, in contradistinction to albumin, are very active in affecting the solubiHty of carbonic acid. This is physiologically important, since serum consequently plays no part 1 The colloid-chemical study of proteins was inaugurated by F. Hofmeister and his pupils; in recent times they have been studied chiefly by Wo. Pauli and his co-workers in numerous experimental investigations. We wish espe- cially to call attention to Wo. Pauli and H. Handovsky, Hofmeister's Beitr. z. Chem. Physiol, u. Pathol., 11, 415-448; Biochem. Zeitschr., 18, pp. 340-371. Loc. cit., 24, 239-262. Further references in the text book of H. Freundlich and Wo. OsTWALD as well as in H. Handovsky, KoU.-Zeitschr., 4 and 5 (1910). ^ Since an absolutely ash-free albumin cannot be prepared bj^ dialysis as shown by the investigations of Pauli and the unpublished experiments of H. Bechhold and J. Ziegler, it is apparent that the question of the electric charge of pure albumin is not yet definitely determined. 148 COLLOIDS IN BIOLOGY AND MEDICINE in respiration. I have no knowledge of researches as to whether the H-ion concentration of water containing CO2 is affected by ash-free albumin. Wolfgang Pauli and M. Samec* have commenced exhaustive studies into the influence of albumins on the solubility of electro- lytes. They employed a serum albumin solution which had been dialyzed eight weeks and contained 2.23 per cent of albumin. All the readily soluble electrolytes investigated showed a slight decrease in solubility as compared with pure water. The solubilities were as follows : Ammonium chlorid Magnesium chlorid Ammonium suphocyanate In 100 gm. serum solution. 27.9 35.51 62.06 Contrariwise, the solubility of difficultly soluble electrolytes was de- cidedly increased by the presence of albumin. The solubilities were as follows: Calcium sulphate Calcium phosphate Ca3(P04)2 Calcium carbonate Silicic acid Uric acid r In 100 gm. water. 0.223 0.011 0.004 0.023 0.040 In 100 gm. serum solution. 0.226 0.021 0.023 0.030 0.057 Having in view the deposition of urates in gout, H. Bechhold and J. ZiEGLER*^ undertook exhaustive studies of the solubility of uric acid and urates in electrolyte-free serum. Since even traces of NaHCOs (in the case of uric acid) and Na salts (in the case of Na-urate) may greatly influence the solubility, before dialysing the serum, HCl was added until the NaHCOs was completely neutra- lizedj and the last traces of Na salts were removed by repeated additions of KCl. Each addition was followed by dialysis. In this way the following solubilities of Na-urate and uric acid were obtained in electrolyte-free serum albumin solution containing 7.6 per cent albumin (expressing the percentage in relation to the entire quantity of protein in defibrinated blood serum) at 37° C. In 1000 gm. serum albumin solution (in 1000 gm. water): Uric acid, 549 to 668 mg 64. 9 mg. Monosodium urate, 476 to 568 mg 1200 to 1500 mg. PROTEINS 149 The ability to decrease the solubility of easily soluble electrolytes and to increase the solubility of difficultly soluble electrolytes is not a specific property of albumins but is common to colloids in general, e.g., gelatin. Albumin and Hydrosols. The exhaustive studies of U. Friede- mann* show that electrolyte-free serum and egg albumin are pre- cipitated both by positive and negative inorganic hydrosols. An optimum precipitation zone exists here, as it does in other colloid precipitations. Excess of albumin or inorganic hydrosol hinders the precipitation. Addition of NaCl shifts the zone of precipitation without, however, conforming to any definite law. As an example I might mention the precipitation (XXX) of albumin by diminish- ing quantities of molybdic acid, with and without added NaCl. Molybdic acid. 0.5 0.25 0.1 0.05 0.025 0.001 0.005 Albumin about 3 per cent. Degree of precipitation in salt-free solution. XXX 0 0 0 XXX XXX xx-xxx Degree of precipitation on the addition of 4 drops of 10 per cent NaCl. XXX XXX XXX XXX 0 0 0 As the result of cataphoretic experiments, U. Friedemann be- lieves that the charge of proteins towards water is not determinative of their precipitation by inorganic hydrosols. Albumin which travels to the anode, notwithstanding this fact gives heavy pre- cipitates with inorganic hydrosols (arsenic trisulphid, silicic acid, molybdic acid). There is much to justify the assumption of U. Friedemann that a given hydrosol, according to its charge, collects at the + or — charge of the amphoteric albumin, thus permitting its aggregation to larger complexes. The albumins appear to act with proteins of definite basic (histones) or acid character just as they do with inorganic hydrosols (U. Friede- mann and H. F.riedenthal*). Influence of Electrolytes. If an electrolyte is added to an amphoteric albumin, the properties of the albumin undergo considerable modification. Salts, even in very small quantities (hundredth normal), raise the coagulation tem- perature, as is shown in the subsequent coagulation temperatures taken from a table compiled by Wo. Pauli and H. Handovskt.*^ 150 COLLOIDS IN BIOLOGY AND MEDICINE Salt 0 0.01 n 0.02 n 0.03 n 0.04« 0.05« NaSCN Na2S04 NaCl NaCaHsOa KSCN 60.3° C. 60.3° C. 60.3° C. 60.3° C. 64.6° C. 68 66.7 63.16 66.9 68.3 69.7 68 65.7 69.2 70.6 68.5 66.4 70.6 69.5 71.6 69.1 67.2 71.5 72.5 69.7 67.9 72.1 70.3 This table shows the remarkable fact that the first traces of salt have a much greater influence than somewhat greater concentra- tions. 0.01 normal Na2S04 added to salt-free albumin raises the coagulation temperature about 6.4° C. while a similar addition to albumin already containing 0.04 normal Na2S04 raises the coagula- tion temperature only 0.6° C. We shall show the significance of this fact later. If the salt is more concentrated, coagulation by heat varies; the coagulation temperature rises continuously in the presence of K, Na and NH4. Thus for 3 normal KCl, coagulation occurs at 75 . 6° C. 3 normal NaCl, coagulation occurs at 73.6° C. 3 normal MgCl2, coagulation occurs at 75 . 4° C. The coagulation temperature reaches a maximum at a certain salt concentration and then falls again in the case of other salts, especially alkaline earths and the allied lithium. Maximum coagulation temperature," C. 6 normal NH4CI 72.8 2 normal (NH4)2S04 74.3 1 normal LiCl 73 . 8 0.5 normal CaCl2 : 71.4 0.5 normal BaCl2 72.2 0.5 normal SrCla 72 Some of the magnesium salts may completely inhibit heat coagu- lation; MgCl2 below 6 normal, Mg(N03)2 below 4 normal. Cations also may be divided into different groups, according to their influence: In the case of SO4, CI, Br and NO3, there is a greater rise with lower concentrations (up to 0.5 to about 1 normal) then a smaller rise up to 1 normal. In the case of SCN and I the inhibition from 1 to 2 normal is so complete that no coagulation occurs even at the highest concentration of the salt. In the case of citrate, acetate and oxalate the coagulation temperature rises sharply from 0.05 to 0.1 normal, whereupon the curve again falls. Obviously this is as- sociated with the strong hydrolytic cleavage of these weak acids in PROTEINS 151 the presence of strong alkalis, whereby there is formed more or less alkali albumin which is not coagulated by heat. We have discussed these questions separately in order that we may obtain a picture of the complicated relations which also reappear in the other properties of albumin. Heat coagulation involves two overlapping processes; albumin be- comes insoluble and it flocculates. Wolfgang Pauli and H. Hand- ovsKY demonstrated this very simply: a mixture of albumin with 2 normal KSCN was boiled and a portion of it dialyzed against run- ning water. The control portion remained clear, but the portion from which the KSCN was removed by dialysis showed marked flocculation. A further influence exerted by neutral salt upon amphoteric al- bumin is the change in viscosity, the internal friction. Although NaCl, NaSCN, Na2S04, CaClz and KSCN in concentrations of 0.01 to 0.05 normal raise the viscosity of water, they depress that of amphoteric albumin solution. If the salt concentration rises, the diminution in the viscosity of albumin may finally be exceeded by the increase in the viscosity of the water, as occurs in fact at 0.1 normal NaCl and (NH4)2S04. Closer observation reveals a far- reaching parallelism between the influence of neutral salts on heat coagulation and viscosity. If non-neutral salts, or salts strongly dissociated hydrolytically, (Na3P04, NaHCOs, AICI3) are allowed to act on amphoteric al- bumin, the result is quite different, since even minute traces of acid or alkali form acid or alkali albumins, which behave quite differently, as we shall see. With higher salt concentration the albumin is salted out or flocculated. Neutral salts of the alkalis as well as inagnesiujji cause a reversible salting out such as occurs also with salts of the alkaline earths, though after a very short time an irreversible coagu- lation sets in. Some of the salts of the heavy metals cause an im- mediate irreversible coagulation. With the alkali salts, the cations (Li, K, Na, NH4) do not materially differ in their salting out action, but the anions do, as may be seen from the following table of F. Hofmeister. The figures refer to the onset of turbidity in egg albumen containing globulin but according to Lewith apply also to ox serum: MoIs per liter at 30-40° C. Sodium citrate 0 . 56 Sodium tartrate 0.78 Sodium sulphate 0 . 80 Sodium acetate 1 . 69 Sodium Chlorid 3. 62 Sodium nitrate 5 . 42 Sodium chlorate 5 . 52 lodid and sulphocyanate do not cause precipitation. 152 COLLOIDS IN BIOLOGY AND MEDICINE Acid Albumin. There is a marked change in the properties of amphoteric albu- min when acid is added to it. It migrates to the cathode as though it were the basic portion of a salt; it loses its coagulability by heat and alcohol, its internal friction is greatly increased and its surface tension diminished. If an excess of acid is added, coagulability by acids and alcohol is restored and its viscosity diminishes. Sjöquist* was of the opinion that albumin formed with acids strongly hydrated (swollen) ionized salts. This assumption was con- firmed by the researches of St. Bugarszky and L. Liebermann* and of K. Spiro and Pemsel.* It was finally established by Mauabe and J. Matuta by extremely accurate measurements on the ioniza- tion constants of acid albumin. W. Pauli and M. Hirschfeld then established that albumin was polybasic, i.e., behaved like a tri- or tetra-amino acid, and that the salts were subject to the normal hydro- lytic dissociation, characteristic of weak bases. S. Oden and W. Pauli conclude from the rise in migration velocity with increasing fixation of acid that polyvalent protein ions are formed.^ In a solution containing about 1 per cent albumin, the maximum internal friction is reached at 0.016 normal HCl, and falls with greater concentrations of acid. Such a maximum is also found with other acids (oxalic acid, sulphuric acid), while with others (acetic acid, citric acid) a continual rise in internal friction accompanies the concentration of the acid. Precipitahility hy alcohol runs parallel with the increase or decrease in the internal friction (K. Schorr). When amphoteric albumin has been made incoagulable by acids, the addition of neutral salts restores the coagulability by heat and alcohol. All the salts investigated (NaS04, NaNOs, Na3P04, Na- acetate, Na-formate, etc.) depress the internal friction. In this re- spect, the cations are of lesser importance, the anions being decisive in the following order: CI < NO3 < SCN < SO4 < C2H3O2. Nonelectrolytes (cane sugar, urea) have, on the contrary, little influence in this respect. Caffein and its salts are an exception, as they increase the internal friction of acid albumin (H. Handovsky*^), An excess of acid alone or the addition of neutral salt to an amount of acid which is insufficient to cause precipitation causes at first a 1 See also W. E. Ringer, Acid Fixation by Albumin and Viscosity, Van Bemmelen-Festschrift (Helder i.H.u. Dresden, 1910), 243-60. PROTEINS 153 reversible flocculation of albumin in the cold, but with greater con- centration (from about 0.03 normal up) an irreversible flocculation. Here also the anions have an unequal influence, which is arranged in an order the reverse of that obtaining for neutral albumin, namely, SO4 < CI < NO3 < Br < SON. This series, accordingly, does not agree with the other one in all respects. It is quite evident in the case of the acid salts that their action is the combined result of the acid albumin formed and the action of the salt itself. The process is, therefore, quite complicated. Alkali Albumin. There is a far-reaching parallelism between alkali albumin and acid albumin. Alkali albumin like acid albumin is not coagulable by heat or alcohol (even 0.003 normal NaOH inhibits the heat coagu- lation of amphoteric albumin) ; its viscosity is greatly increased, its surface tension diminished; excess of alkali restores the precipitability by alcohol and again decreases the internal friction; it migrates to the anode. St. Bugarszky and L. Liebermann* showed that NaOH was bound by albumin, and that albumin depressed the freezing point of soda-lye. Neutral salts arrest the action of alkalis; in contradis- tinction to acid albumin it is the cations to which the greatest signifi- cance attaches, and, in fact, the effect of the divalent earth alkalis (Ca, Sr and Ba) and the divalent magnesimn very greatly exceeds that of the monovalent alkalis. Though heat coagulation does not occur at all or advances only to a milky turbidity (e.g., the effect of 1.2 normal KCl was doubtful), in alkali albumin containing large quantities of alkaline salts the ability of alkali albumin to coagulate with 0.003 normal NaOH is demonstrable upon the addition of 0.0002 normal CaClg. Additions of neutral salts bring about a decrease of internal friction in a manner analogous to their influence on heat coagulation, and, in fact, a small addition of salt has a proportionately greater effect than a large one. Moreover, the earth alkalis greatly exceed the alkali salts in their ability to diminish internal friction. The salting out of alkali albumin requires a greater concentration of alkali salts than is required for neutral albumin; the product is reversible and the anions are effective in the same order as for neutral albumin. In general, the relations are simpler for alkali albumin than for acid albumin. In the former, they depend upon the electrolytic dis- 154 COLLOIDS IN BIOLOGY AND MEDICINE sociation of the base, while in the latter, certain electrochemical factors which may not be disregarded play a part. If dilute soda-lye (0.025 normal NaOH) acts for a long time on serum albumin, the internal friction reaches a maximum, remains constant for a while and then diminishes (K. Schorr). Evidently there occurs fixation of water, swelling. The cleavage of the albu- min molecule is accompanied by the formation of less colloidal disintegration products, and is characterized by a diminution of the viscosity. If from these results we try to obtain an idea of the processes in- volved, we shall find a useful guide in the theory of the amphoteric nature of genuine albumin proposed by G. Bredig* and extended by Wo. Pauli. Let us think of albumin as built according to the structure of a cyclic ammonium salt : R ^COO in which R represents a complicated organic complex and the ab- sorption of water follows according to the scheme: yNHs .NH3OH R + H.2O <=± R ^COO ^COOH This is an amphoteric electrolyte which unites with bases and acid, which splits off H as well as OH ions and in which the Ka (acid dissociation) > Kb (base dissociation) in other words, it behaves like a very weak acid. Pure albumin consists principally of electrically neutral particles but forms acid and alkali salts which are strongly ionized. There exist xNHaOH ^NHsCl ^NHsOH R R R ^COOH ^COOH ^COONa neutral albumin acid albumin alkali albumin That the albumin ions are responsible for the great internal friction is to be assumed from the investigations of E. Laqueur and 0. Sackur* on alkali-caseinates. The cause of this phenomenon is found in the strong hydration (water fixation, swelling) of the albu- min ions. According to Wo. Pauli and M. Samec the existence of polyvalent ions must be assumed in the case of acid and alkali albu- PROTEINS 155 min. Even assuming the smallest values for the molecular weight of albumin, the quantities of acid or alkaU found are so large that they indicate the fixation of several acid or alkali molecules. This offers a further explanation of the marked increase in hydration produced by acids and alkalis. The stability of an albumin solution and its precipitability, e.g., by alcohol, are directly proportional to the num- ber of albumin ions it contains. The circumstances here are quite analogous to those with crystalloids. Ions tend to go into solution and to form hydrates; the saturation concentration of neutral par- ticles is always less than that of ions. In this way, we may explain the properties of strongly ionized pure acid and alkali albumin as contrasted with the slightly disso- ciated neutral albumin. How does this theory agree with the effect of neutral salts? "VYo. Pauli explains it in the following way: ,NH4C1 yNHaCl R + NaNOa i^ R + HNO3 ^COOH ^COONa Acid albumin + neutral salt of acid^lbllmin + ^^ee acid In this way was explained not only the increased number of free H ions, which he demonstrated, but also the marked diminution in internal friction; because an amphoteric salt, in which both anions and cations tend to ionize about equally, is but slightly dissociated. The action of neutral salts in alkali albumin is different; it follows the following scheme: yNHaOH .NH2KCI R + KCl ?=± R +H2O ^COONa ^COONa Alkali albumin + neutral salt complex^albumin ^ ^^^^^ Accordingly, a complex albumin salt was formed to which a less amount of ionization may be ascribed than to alkali albumin. The action of salts of the alkaline earths follows this scheme: .NH3OH .NHaNaNOs R + ^ NO3 :^ R + H2O ^COONa ^COO^ The replacement of the alkali ion in the carboxyl of the amino group results in a weakly ionized complex salt. The effect on albu- min of organic bases, which are often highly toxic, and of amphoteric electrolytes, have also been studied by H. Handovsky, and the re- sults agree with the above scheme. 156 COLLOIDS IN BIOLOGY AND MEDICINE The conditions governing the action of neutral salts upon acid albumin are not sufficiently understood to warrant proposing a simple scheme.^ The optical rotation of albumin runs parallel with the changes in its internal friction and coagulability (Wo. Pauli,*^ M. Samec, E. Strauss). In fact, the albumin ions rotate light more power- fully than neutral albumin. Let us summarize briefly : neutral albumin has a low internal fric- tion, coagulates easily and shows little optical rotation; ionized albumin has high internal friction, coagulates with difficulty and rotates light powerfully; neutral salts diminish ionization. This chemical point of view is additionally supported by the investigations of P. Pfeiffer and J. W. Modelski as well as of P. Pfeiffer and Wittka. These authors have shown that amino acids and Polypeptids of known chemical structure form with neutral salts of the alkalis and earth alkalis, crystalline addition compounds constructed on simple stoichiometric principles. Some of these molecular compounds are much more readily soluble in water than the aminoacids or Polypeptids and some much less soluble, so that, as in the case of albuminous substances, it is possible to salt some of them out (analogous to globulins). Albumin and Inorganic Hydrosols According to U. Friedemann*^ electrolyte-free albumin is precipi- tated both by positive and by negative inorganic hydrosols. Hydro- phobe hydrosols such as AS2S3, Au, etc., regularly form precipitates, which, according to W. Pauli and Hecker, are not inhibited by an excess either of hydrosol or of albumin. Neutral salts, acids and alkalis exert a protective action, but nonelectrolytes, such as urea and sugar, are inactive. In the case of positive hydrophile inorganic hydrosols such as Fe (OH) 3 there is an optimum precipitation zone that lies somewhere between one part by weight of Fe (OH) 3 and three parts by weight of the electrolyte-free albumin. With an excess of Fe(0H)3 there is increasing solution which is complete in about the proportion of two to three; there is no complete solution with an excess of albumin. Neutral salt exerts a protective action when albumin is in excess but on the contrary favors precipitation when Fe (OH) 3 is in excess. Acids inhibit precipitation; alkalis precipitate when Fe (OH) 3 is in ^ From the formula it should not be assumed that only free terminal NH2 groups are considered. As the result of the work of Blasel and J. Matuta on deaminized glutin (glutin whose free NH2 groups are satisfied) it is more probable that its interior NH groups are involved in the formation of salts with acids. PROTEINS 157 ..•■ excess, otherwise they exert a protective action. Hydrophile negative inorganic hydrosols, e.g., siHcic acid, differ from positive hydrosols only by the presence of H and OH ions which act oppositely to those in the positive hydrosols. Only a small fraction of the albumin is precipitated by hydrophobe inorganic colloids; but the greater portion, and at times all the albu- min, is precipitated by hydrophile hydrosols. Albumins appear to react with proteins of pronounced basic (histone) or acid character (U. Friedemann and H. Friedenthal*) just as do inorganic hydrophile hydrosols. Albumin, Heavy Metals and Salts of Heavy Metals. On shaking salt-free albumin solutions with metallic iron, cobalt, copper, lead, nickel or aluminium, portions of these metals go into solution and are bound by albumin in a hitherto unrecognized "masked" form, according to Benedicenti and Revello-Alves. Electrolyte-free albumin yields no precipitate with zinc, copper, mercury or lead salts. In the presence of salts, however, albumin forms with salts of the heavy metals precipitates whose chemical com- position is not constant, but depends on the concentration of the components at the time of precipitation. By precipitating albumin with solutions of heavy metal salts of varying concentrations we get ''irregular series," which frequently show two zones of precipi- tation: one with very dilute solutions of the metal salt (one ten thousandth normal and under) and another with high concentration; between these there is always a zone with no precipitation. The precipitation zone with great dilutions of the metal salt is due accord- ing to H. Bechhold* to metal hydroxid split off hydrolytically, which precipitates with albumin, forming an insoluble heavy metal- albumin compound. The resolution of this precipitate at somewhat greater concentration of metal salt results from ionization. W. Pauli and Hecker have shown by very convincing experiments upon the action of FeCls on albumin that a soluble ferric ion-albumin complex occurs somewhat in accordance with the following scheme [a;Fe(OH) 3 -protein] + gFeCls = [xFe(OH) 3- protein] gf Fe + 3gCl. Upon addition of more FeCls, just as when acid is added to acid albumin, partial neutralization occurs and there is further precipi- tation. Ur02Cl2 behaves hke FeCls, as do also, to a certain extent, AgN03, ZnS04 and Pb(N03)2. On the other hand, the precipitate disappears with higher concentration of CuCi2 and HgCl2, and, abso- lutely no precipitate is formed with electrolyte-free albumin and the chlorides of Fe", Co", Mn", Cd". 158 COLLOIDS IN BIOLOGY AND MEDICINE Globulin. Those proteins which are insoluble in pure water and soluble in salt solutions are called globulins. They are constituents of the blood serum^ eggs and milk of animals. They occur in other organs in traces, thus, e.g., thyreo-globulin, the iodin-containing protein of the thyroid, is a globulin. Large quantities of globulin are found stored in the seeds of plants. A seed globulin, edestin, has been obtained in crystalline form. If serum is dialyzed against pure water, globulin will be precipitated as the content of the dialyzer cell (globulin) parts with salt. By ultrafiltration, H. Bechhold*^ was able to separate globulin from the common salt holding it in solution. Globulins are also soluble in acids and alkalis. If globulins are kept undissolved (e.g., dried or sus- pended in distilled water) a change occurs ; they lose more and more of their solubility in dilute solutions of neutral salts. Like the albumins, globulins are amphoteric: without the presence of salt they have no definite direction of migration; in the presence of traces of alkali they pass to the anode and in the presence of acids they pass to the cathode. According to L. Michaelis, an H ion concentration of 4.10"'^ is the isoelectric point for serum albumin. According to W. B. Hardy, *^ a given quantity of salt-free globulin is dissolved by an equimolecular quantity of strong monobasic acids (HCl, HNO3, monochloracetic acid). The weaker the acid the more of it is neces- sary to dissolve the globulin. About twice as much sulphuric acid, tartaric acid and oxahc acid, and three times as much phosphoric acid and citric acid, is required than of HCl. W. B. Hardy concludes from this that globulins form salts with acids which in the case of weak acids are greatly hydrolyzed. Bases act in a manner similar to the acids, with the exception that NH3 dissolves as much globulin as NaOH. Rise of temperature increases the hydrolysis, i.e., globulin, dis- solved in an amount of weak acid or weak alkali just sufficient to give a clear solution, becomes turbid when it is warmed; however, the process is not completely reversible. It was deduced from the conductivity values of alkali globulin that globulin is a pentavalent acid, and from its saponification with methyl acetate as well as its action in the inversion of cane sugar, that it is of a more strongly acid than basic character. This is also evident from the fact that the conductivity of its acid salts increases progressively more, when diluted, than the conductivity of its alkali salts. The preponderant acid character is also evident from the fact that litmus is reddened by globulin. PROTEINS 159 .J As in the case of albumin, the globulin ions are responsible for the internal friction. Though the internal friction of globulin in neutral salts is low, it is considerably higher in the ionized solutions occurring in acids or alkalis; the viscosity is highest in the case of alkali salts of globulins which are ionized most strongly and least hydrolyzed. The viscosity rises disproportionately with concentration and, in fact, the increase for alkali globulin > for acid globulin > for neutral salt- globulin (W. B. Hardy*2). W. B. Hardy gives the following viscosity values for 7.59 gm. globulin per liter: Water 1 MgS04-globulm 4.66 HCl-globuUn 15.5 NaOH-globuUn 67.9 He derived these velocities for globulin ions: Acetic acid-globulin 23 • 10~^ cm. per second HCl-globuUn 10-10^ " " NaOH-globuUn 7.7-10-^ " " " W. B. Hardy regards solutions of globulin in neutral salts as molecular combinations, since, in contrast to solutions in alkalis or acids, they are thrown down upon dilution. It can be understood from the dominant acid character of globulins that a neutral salt solution of globulins is precipitated by acids. Though alkali globulin solutions are permanent in the presence of neutral salts, acid globu- lins are precipitated by them. According to W. B. Hardy, serum contains no globulin ions. If serum is kept warm for a long time {e.g., 2 hours) below its coagulation temperature, the amount of globulin is increased at the expense of the albuminous portion (Moll*). This formation of globulin is either impeded or entirely stopped by salts. "Artificial globulins" is the designation of the substances pre- pared from egg albumin by Andre Mayer.* He found that when he added to egg albumen a certain quantity of a solution of a salt of a heavy metal ZnS04, Zn(N03)2 or a positive colloid (colloidal Fe203), the resulting precipitate was insoluble in water and in solutions of nonelectrolytes, but, on the contrary, it was soluble m solutions of salts (e.g., NaCl, Ca(N03)2, etc.). With these tacts in mind, we must consider the suggestion made by A. Mayer that globulins are com- plexes of albumins (possibly with other positive colloids). 160 COLLOIDS IN BIOLOGY AND MEDICINE Fibrin. Fibrin is the substance of blood plasma, which coagulates shortly after the blood has left the vessels. Upon the clotting of plasma, which contains no blood corpuscles, no jelly is formed, but character- istic fibrous masses. Formerly it was thought that uncoagulated fibrin, called fibrinogen (see p. 299), was something quite different from fibrin. As a result of the investigations of Hekma it is possible that fibrinogen is the hydrosol of alkali fibrin. If fibrin is dissolved in extremely dilute alkali we obtain a fluid having all the properties of fibrinogen. Normal coagulation outside the blood vessels as well as the resulting product must be sharply differentiated from fibrin coagulated by heat. Fibrin coagulated by heat ceases to show the swelling phenomena it possessed before it was heated; it has become hydrophobe. When coagulated, fibrin is an irreversible gel. In weak acid and alkalis it swells and gradually goes into solution following, as it does so, the same laws as does gelatin (see p. 68, et seq.). Martin H. Fischer* has studied its swelling under the in- fluence of acids, bases and salts, and utilized his results for his theory of edema (see p. 223, et seq.) Muscle albumin or myosin, the coagulation of which at death causes rigor mortis, belongs to the same group as fibrin. Nucleins. Basic substances have been prepared from cell nuclei; histone from the leucocytes of the thymus, fish roes, etc., as well as pro- tamine, so thoroughly studied by A. Kossel and usually obtained from the spermatozoa of several different kinds of fish. They do not exist as such in these organs but occur in combination with acid nucleins as nucleo-proteins and nucleo-histones. Neutral solutions of histone yield a precipitate containing very little salt with solutions of egg albumin, casein and serum globulin. When we recall that casein and globulin are of decided acid reaction, their union with basic histone is quite easily understood. A priori, it is improbable that the precipitate should contain 1 part histone, 2 parts casein and globulin and 1 part egg albumin, as has been claimed. It has been shown by U. Friedemann and H. Frieden- thal* that according to the relative concentration in which solutions of histone and albumin are mixed, the precipitate will vary in com- position; that the addition of NaCl changes the precipitation limits and that fresh solutions have different precipitation limits than older ones. All these facts point with certainty to the fact that nuclein is not a definite chemical combination, but that nucleins are colloid compounds consisting of a negative and a positive colloid. PROTEINS 161 Albuminoids. (Scleroproteins) Though the organic framework of plants consists of cellulose, that of animals is formed of nitrogenous substances classified as albuminoids. Like cellulose, they are very resistant chemically to foreign influences, water, salt solutions, acids and bases. The most important of the albuminoids is collagen, derived from bone, cartilage and the fibrils of connective tissue. On boiling with water, it swells and gradually dissolves, undergoing hydrolytic cleav- age and forming glue or gelatin. Gelatin, which has been the subject of the most important investigations concerning hydrophile gels and from which the whole class of gels take their name, does not occur in the organism at all. The most important data concerning it have been given on page 68, et seq. What has been said, especially in reference to the preparation of a solution of agar (p. 137) holds for gelatin as well. It should be recalled that acids and alkalis greatly increase the swelling of gelatin. The swelling capacity reaches a maximum with increasing concentration of HCl (0.025 n) and KOH (0.028 n) (Wo. Ostwald). We thus find an absolute parallelism between the swelling of gelatin and the ionizati(?n of albumin (see pp. 152 to 156). In excellent agreement with this is the fact that the minimal swelling occurs at the isoelectric point of gelatin, namely, with an H ion concentration of 2.10"^ (L. Michaelis, R. Chiari). It must be emphasized especially, that a very dilute solution of gela- tin depresses (according to G. Quincke) the surface tension of water 12 per cent. The solubiHty of CO2 is very considerably greater in gelatin sols than in water (in contrast to other hydrophile sols). Compared with other colloids (serum albumin), gelatin lowers the solubility of easily soluble electrolytes and increases that of those soluble with difficulty. The following are the figures from the in- vestigations of Wo. Pauli and M. Samec:* There dissolves in 100 gm water + i per cent gelatin + 10 per cent gelatin Ammonium chlorid 28.49 27.55 26.48 Magnesium chlorid 35.94 35.22 35. 13 Ammonium sulphocyanate 62.46 61.46 58.92 1.5 per cent gelatin Calcium sulphate 0. 223 0.295 Tertiary calcium phosphate Ca3(P04)2 0.011 0.018 Calcium carbonate 0.004 0.015 Silicic acid 0.023 0.027 The solidification and the melting points depend greatly upon the previous history of the gelatin; the longer gelatin is warmed the less it tends to solidify. Upon heating a 2 per cent gelatin solution 162 COLLOIDS IN BIOLOGY AND MEDICINE to 100° C, the relative internal friction (according to P. von Schroe- der) falls from 1.75 (at the end of one-half hour) to 1.22 (at the end of 16 hours). Possibly this is due to the increasing hydrolytic cleav- age. The following figures give some idea of the relations: Content per liter. Solidification temperature, ° C. Melting temperature, ° C. Grams. 1.8 2.5 50 100 150 <10 (Rohloff and Schinja) 0 (S. J. Levites) 17.8 (Pauli and Rona) 21 (Pauli and Rona) 25 . 5 (Pauli and Rona) 26.1 (Pauli and Rona)* 29.6 (Pauli and Rona) 29.4 (Pauli and Rona) These solidification temperatures are markedly shifted by elec- trolytes and, in fact, the anions have the greatest influence, whereas the cations are of less moment. The solidification temperature is raised by ) SO4 > CH3CO2 > The solidification time is shortened by ) tartrates. ^, Ti-r. .• ^ . • , n ibenzoates and salicy- The solidification temperature IS lowered by , ^ ^ o/^tvt^ t^ -n rru vA-ti V +• • 1 +V. Au \ lates>SCN>I>Br The solidification time is lengthened by .^ ^^^ ^ r^^ ^ -^ ] > NO3 > CI. The following data (from H. Bechhold and J. Ziegler*^) serve as an example: Melting point. 10 per cent gelatin 31.6 10 per cent gelatin + 1 mol. NaCl 28. 5 10 per cent gelatin + 2 mol. Na2S04 34. 2 10 per cent gelatin + 1 mol. Nal 10.0 Nonelectrolytes also influence the melting point of gelatin. Glycerin and sugar (mannit, cane sugar, etc.), in contradistinction to agar, raise the temperature and increase the rate of gelatinization, while furfurol, urea, alcohols, resorcin, hydrochinon and pyrogallol lower them. Nongelatinizing colloids have no influence on gelatini- zation. The following figures from H. Bechhold and J. Ziegler*^ serve to make this clear: Melting point. 10 per cent gelatin 31 . 66 10 per cent gelatin + 1 mol. grape sugar 32 . 25 10 per cent gelatin + 2 mol. glycerin 32. 17 10 per cent gelatin + 2 mol. alcohol 30 . 0 10 per cent gelatin + 1 mol. urea 26. 3 Precipitation of the gelatin sol must be sharply differentiated from gelatinization. Precipitation is induced by electrolytes, whereas nonelectrolytes usually interfere with it. Precipitation corresponds PROTEINS 163 rather to salting out, which, we may assume, occurs also in the case of crystalloids; it may be induced not only by electrolytes which raise the melting point of the gel, but also by those which depress it. Precipitation becomes evident at first through a turbidity which may be sufficiently marked to give a tenacious gelatin phase and a more limpid aqueous phase. In precipitations, also, the anions have the determining influence, their precipitating effect being arranged in the following order: SO4 > Citrate > Tartrate > Acetate > Chlorid. Inorganic hydrosols behave quantitatively toward gelatin the same as toward albumin (see p. 156). The swelling and shrinking of gelatin referred to on page 68, et seq., are characteristic for all elastic gels. According to J. Traube and F. Köhler there exists a parallelism between the swelhng, shrinking, solidification and melting point of gelatin when it is mixed with other substances. The tinctorial properties of elastic fibers and their chief constituent, elastin, are better known than their other colloidal properties. To the investigations of P. G. Unna and L. Golodetz, we owe our knowledge of the keratins, the horny substances composing skin, hair, nails, hoofs, horns, feathers, etc. Chemical studies of these substances are very difficult because they resist chemical attack. We shall merely mention the remaining albuminoids, s-pongin, the structural support of ordinary sponge, chonchiolin, the framework of mussels and snails and further, the albumoids, a group into which almost all unclassified proteins are thrown. Nucleoalbumins. These proteins, like the albumins, are digested by pepsin-hydro- chloric acid; they dissolve almost entirely, but at the same time split off an almost insoluble phosphorus-containing complex. The casein of milk, the vitellin of egg yolk and perhaps also legumin and vegetable casein are nucleoalbumins. It is remarkable that among the phosphorus-containing proteins obtained from seeds, there should be several that are soluble in alcohol (gliadin from cereal grains and zein from corn). Because of this, it is a question however, whether they are related to casein. On account of its importance, casein has been most extensively investigated. In milk, casein exists as a salt (united to lime and alkali) and, being dissolved, exhibits profound hydrolytic dissociation. Casein may be thrown out of solution by the addition of acids or ren- 164 COLLOIDS IN BIOLOGY AND MEDICINE nin, yet the casein obtained by the addition of acids and that obtained by rennin are not identical. Furthermore, casein may be separated from the crystalloid portions of milk by ultrafiltration (H. Bechhold and by centrifugation (H. Friedenthal*). We are indebted to E. Laqueur and 0. Sackur* as well as T. B. Robertson* (who gives a bibliography) for the exhaustive chemical studies of casein upon which we base our remarks. In water, casein is completely insoluble and decidedly acid. A piece of casein stains damp blue litmus paper red. According to L. Michaelis the isoelectric point occurs when the H ion concentration is 2.10"^. Casein forms salts soluble in water with alkahs and alkaline earths. One grain of casein binds 8.81 c.c. 1/10 normal alkali (using Phenolphthalein as an indicator). From this the combining weight of casein is 1135, and a common multiple of this is its molecular weight. E. Laqueur and 0. Sackur deduced from the conductivity of sodium-casein solution with increasing dilution, that casein was a tetra- or hexabasic acid and that, therefore, its molecular weight lay between 4540 and 6810. T. B. Robertson, as a result of his investigations, comes to the contrary conclusion, that only a single carboxyl group is available for union with a base. W. van Dam* has investigated the diminution in H ion concentration of lactic acid solution upon adding casein and concludes from it, that a basic group unites with four replaceable H atoms in a casein mole- cule. In solution, casein salts are hydrolytically dissociated and, in fact, it follows from the following experiment that casein (acid) forms a hydrosol. A neutral solution of casein-sodium solution is slightly opalescent and becomes clear upon the addition of an alkali. The solution of casein-lime salts are still more opalescent since the alkaline earth salts are weaker bases. Casein-sodium does not diffuse through parchment; the membrane must have a decided difference in potential since the sodium ion has a strong tendency to diffuse. E. Laqueur and 0. Sackur showed that the internal friction of casein salt solutions increased proportionately to the electrolytic dissociation, and that every diminution of electrolytic dissociation was accompanied by a diminution of internal friction. The casein ions are thus responsible for high internal friction. Hemoglobin. Hemoglobin, the coloring matter of blood, has only recently been studied by P. Bottazzi. It is preeminently suited for colloid-chemi- cal investigation on account of its color, ease of crystallization and PROTEINS 165 pronounced colloid character. Chemically, it is composed of the protein glohin, a histone, and the iron-containing component, hematin, which is apparently a pyrrol derivative. Bechhold used 1 per cent hemoglobin solutions to gauge his ultrafilters (see p. 99). After dialyzing three to four months, hemoglobin solutions have a conductivity of K 20° = 1 X 10~*. After dialyzing five and a half months, the hemoglobin was completely precipitated though the pre- cipitate did not have the amorphous flocculent character of other pro- teins but was more granular although no crystalline formations could be recognized. If the granules were removed by filtration during the dialysis, there was obtained a reddish, optically inactive solution which showed no particles in the ultramicroscope. Such a solution does not pass through the dialyzing membrane and contains particles which are somewhat larger than those of serum albumin as deter- mined by ultrafiltration. Regarding the absorption of O and CO2 by hemoglobin see page 308, et seq. During dialysis, hemoglobin changes to methemoglobin. Methe- moglobin, which is insoluble in water and neutral salts, redissolves upon the addition of traces of alkalis or acids. Purified hemoglobin migrates to the anode. In view of this fact and the relatively high conductivity of a dialyzed hemoglobin solution P. BoTTAZzi assumes that hemoglobin is an hemoglohinic acid in- soluble in water, but which exists in solution as an alkali hemoglo- binate. Being an amphoteric electrolyte it is also soluble in acids and then migrates to the cathode; its H ion dissociation far exceeds its OH ion dissociation. According to L. Michaelis, on the contrary, hemoglobin is less acid than serum albumin; its isoelectric point occurs with an H ion concentration of 1,8.10"''. The viscosity curves which P. BoTAZzi*^ obtained on dissolving it in alkalis and acids indicate the occurrence of methemoglobin ions; they resemble the viscosity curves of alkali and acid albumin. Completely dialyzed methemoglobin coagulates at 47°-53°C.; in the presence of traces of alkalis or acids, and in the absence of neutral salts coagulation fails to occur even at 100° C. In conclusion we shall mention the mucins and mucoids. They are the excretory products of many glands and may be briefly de- scribed as animal mucus. The possession of a carbohydrate in ad- dition to the protein component distingufshes them chemically. CoUoid-chemically they also occupy an intermediate position, since they are not coagulated by heat but are precipitated by salts and alcohol. Because of their acid character they are precipitated by acids and dissolved by alkalis. 166 COLLOIDS IN BIOLOGY AND MEDICINE The Colloid Cleavage Products of Proteins. Proteins undergo hydrolytic cleavage by acids, alkalis and enzymes. From the viscosity curve resulting from the prolonged action of NaOH on albumin, it is deduced that the disintegration of albumin depends upon the albumin ions (W. M. Bayliss, K. Schorr). T. B. Robertson arrived at a similar conclusion from his studies of the tryptic digestion of casein. Under these circum- stances, it is obvious that, in general, the digestion of albumin by enzymes occurs more readily in acid or alkaline solution than in neutral solution where there are but few albumin ions. With increasing subdivision of the molecule, the diffusibility, etc., increases, and the precipitability by neutral salts decreases. The group of cleavage products, known as alhumoses, dialyze slowly through animal membranes, whereas the peptones are not to be dis- tinguished in this respect from true crystalloids. That they are still to a certain extent colloids is proved by their forming films on the surface of water (described on p. 33) which places them in the class of those dyes which lie midway between colloids and crystal- loids. H. Bechhold arrived at a similar result by separating albu- moses from the remaining fluid by means of ultrafiltration, whereas peptones and the closely related deuteroalbumoses C were not held back even by 10 per cent filters. The albumoses and perhaps also the peptones are evidently mix- tures of numerous different substances which have not yet been chemically identified. They are differentiated and classified ac- cording to their precipitability by electrolytes and alcohol, which doubtless stands in a certain relationship to the size of their molecules and particles and to their ionization, see page 152. By means of ultra- filters of different permeability, H. Bechhold separated albumoses into "various groups which corresponded to their precipitability. Subjoined is the classification of F. Hofmeister-Pick of the results by ultrafiltration: Hetero- and protoalbu- moses Deuteroalbumoses A . . . Deuteroalbumoses B . . . Deuteroalbumoses C. . . Peptones Saturation per cent of ammonium sul- phate required to produce precipita- tion. J 24-42 54-62 70-95 100 + acid Not salted out Ultrafilter, 2.5 per cent. 3 10 (H4) 10 Saturation per cent of am- monium sulpliate required to produce precipitation. Residue 23 Filtrate 34 Residue " 34 Filtrate 95-100 Residue 95-100+ acid, filtrate, small frac- tion Pass through filter PROTEINS 167 These experiments were undertaken and largely carried out by Edgard Zunz.*^ They lead to a multitude of valuable results. It suffices to mention that by ultrafiltration of thioalbumose/ there were shown to be two components of obviously different chemical constitution; and that in hetero- and protoalbumose at least two and in the latter probably even three ''proteoses" should be assumed. ^ Deuteroalbumose A, on account of its high content of easily spht-off sul- phur, is termed " Thioalbumose " by Pick. CHAPTER XI. FOODS AND CONDIMENTS. Formerly the preparation of food was one of the most important tasks assigned to the housewife; nowadays among the middle and better classes this duty is almost entirely surrendered to servants, while among the working classes the women can give it but little at- tention as they must increase the family income by work away from home. These conditions have brought with them the steady de- terioration of the Art of Cookery. The raw materials nowadays supplied to the kitchen from wholesale establishments, e.g., the bread, fruit, vegetables, beer, and perhaps even meat, etc., are, it is true, of a much superior quality than formerly. This is due to com- petition, easier means of communication, improvement in methods of cultivation and all the advantages consequent upon production on a large scale. The conversion of this raw material into palatable meals requires a large measure of experience, loving care, and great interest — which one can expect from neither a twenty-year-old cook nor the tired working woman. Nutrition is undoubtedly the most important factor in our whole social life; if we place the yearly expense for nourishment in the German Empire at ten milhards of marks, it is surely underestimated. Only a one per cent increase of the successful utilization of food would show a yearly profit of at least one hundred million marks ($25,000,000). It is hardly to be expected that we shall accompUsh this by re- turning to former conditions, but rather, in a different way, namely, the development of the art into the science of cookery. The kitchen will probably adapt itself more and more to wholesale prepara- tion, and then there will be men and women who will choose cooking as a profession because of their scientific training for the work. Colloid Chemistry furnishes us the rules for the selection and preparation of our foodstuffs; because cooking is nothing but prac- tical colloid chemistry. Our foodstuffs consist entirely of colloids and their nutritive value is to be judged mainly from a colloid-chemi- cal point of view. 168 FOODS AND CONDIMENTS 169 Very little truly scientific work ^ has as yet been accomplished in this field; so that we must content ourselves with indicating the problem.* Meat. What we consume as '' meat " consists for the most part of muscle fibers and connective tissue with the interspersed fat. In judging the meat of healthy animals its source is the chief criterion; young, well-nourished animals possess a juicy meat and tender con- nective tissue, whereas old worn-out animals are less juicy and their connective tissue shows a firmer structure. From these few premises it is evident that it is a question of turgor and swelling capacity, which change with age; this is an important colloid-chemical prob- lem. It is still an open question whether the toughening of con- nective tissue may be compared to the lignification of the vascular bundles of plants, which according to H. Wislicenus* (see pp. 249 and 250) is due to the adsorption of colloids from the cambial sap. Fresh killed meat is tender; it becomes soft again only upon the disappearance of rigor mortis. This is conditioned by phenomena of swelling and shrinking, an interesting method for studying meat adopted by 0. von Fürth and E. Leuk. After death lactic acid accumulates in the muscular tissue, and greatly increases the swell- ing capacity of the muscles (see p. 290). If such a muscle is placed in a dilute salt solution it swells up and after about 25 hours has taken up a maximum amount of water of swelling. Then shrinking occurs as the result of the progressive coagulation of the muscle albumin. The curve obtained in this manner has a quite characteristic shape depending on what has happened to the meat. Fig, A shows the curve of swelling of horse heart three or four hours after slaughter; Fig. B, after it has been kept 3 days in the ice 1 In this connection Dr. J. G. M. Bullowa mentions favorably "The Chemistry of Cookery," by Mattietj Williams, London, Chatto & Windus, 1892. [H. C. Sherman has included in his book, "Food Products," Macmillan Co., 1917, the important recent data. By reason of the war, the preservation of perishable food- stuffs has assumed great importance. Dehydrating processes make possible the transportation of large quantities of vegetables in limited cargo space. Clarence V. Ekroth has presented an excellent account of the methods and Uterature of dry- ing and dehydrating foods in Allan Rogers' " Manual of Industrial Chemistry," Van Nostrand, 1918. The products dried by Ekroth's " G. H." Evaporator in the Mrs. Oliver Harriman's Food Research Laboratory are of excellent quaUty; are said to retain the important food accessories. The " G. H." Dehydrator dries its product in moist air at a moderate temperature. Humidity and temperature are controlled. A fan blower is provided for recirculation of the air so that volatile substances are kept in contact with the product and only a small per- centage is lost. The process is quite the reverse of ordinary cooking when an effort is made to retain moisture and volatile substances by rapidly seahng the surface of the food by quick heat coagulation. Tr.] 170 COLLOIDS IN BIOLOGY AND MEDICINE box. In the first instance it absorbs approximately 10 per cent of its weight in the first 25 hours, then shrinking occurs. The ice box heart, on the contrary, immediately begins to lose water, and at the 20 4.0 50 60 30 Hours Swelling of Horse Heoir+ 3or4 Hours, Post Mortem Fig. a. end of 45 hours has lost by shrinkage about 55-75 per cent of its water. Typical are curves, Figs. C, D, E, which show comparatively zo tlO 20 «51 c 30 U 40 50 60 ~ ■" ~" — — — r] ~ — ~ — — ~ ~ ^a L^ \ V \ V s N \ \ \ > N \ *v ^ V ■»^ \ ^ V ^ ~> > y \ - "" — — ,.^ ■- — ._ 0 1 0 £ 0 3 0 A 0 5 0 6 0 Hours Swelling of Horse Heart after Cold Storage for 3 or 4 Hours Fig. B. butcher's meat, cold storage meat and hare, which have been kept a year at — 10°. In actual practice the method is to weigh morsels of meat of as nearly the same size as possible, and after placing them in salt solution, at hourly intervals, to determine the changes in weight. FOODS AND CONDIMENTS 171 Percentage Change in Weight I + _ uit^wro — o — row*- ui g« -^ o ^0><^0 0 -x^ '~~~ h- — 0 3 v-^ ^X" \ 0 ^ y 0 / Swelling of Hare Flesh after Preservation for more than a Year at -10° C Fig. E. 172 COLLOIDS IN BIOLOGY AND MEDICINE Naturally a salt solution of the same concentration (5-10 per cent) is always employed. In pure water, shrinking occurs immediately, since muscle albumin coagulates in it spontaneously. High concen- trations of salt, 25-30 per cent, likewise depress the curve. New problems arise in cooking meat. If meat is boiled in pure water we obtain a "weak" broth. Muscle albumin coagulates in water before heat coagulation occurs and this impedes the exit of the crystalloid. Salt is therefore added immediately if good soup is to be expected. With hoiling, heat coagulation occurs whereby the meat loses from 20 to 30 per cent of its water. The mechanism of this loss is still unknown. We can understand why there should be a loss of from 20 to 35 per cent in roasting, and it would be still greater if the surface were not constantly protected by pouring over or dipping into fat (basting). Preserved meats are less perishable because they contain less water and because the muscle albumin has been converted into a character- istic gel condition. This end is attained in various ways: In pick- ling, water is removed from the meat by the salts of the brine, while at the same time there is an exchange of crystalloids, whereby salts enter from without which change the albumin as regards its coagulability and sweUing capacity; and extractives leave it and are removed with the brine. Of course very important changes occur during storage, so that according to A. Gärtner, with in- creasing age pickled meat becomes more difficult of digestion and loses 30 per cent of its nutritive value. Smoking of meat is usually preceded by a short pickling process. The abstraction of water in this case is accomplished by means of a strong current of air, and in the dried meat (pemmican) which is much rehshed in some regions, in the Arctics (for instance) , there is no loss other than water. Naturally, the properties of every gel, materially, depend upon its history. To quote a single example : F. Stoffel* (in the laboratory of Prof. H. Zangger) found, that the diffusibility of one and the same substance through the identical gelatin differed, depending upon whether the gelatin was rapidly solidified with ice or slowly cooled at room temperature. Accordingly, we may assume in the case of meat, that the properties of the coagulated albumin will vary with the conditions maintained during coagulation, and that upon these depends its food value. An essential question in the investigation of sound meat, pre- served meats and food preparations must be their available food value, which can be answered only by complicated and expensive metabolism investigations. From my point of view this ought to be a fruitful field for the colloid chemist, who ought certainly to be FOODS AND CONDIMENTS 173 in a position to replace with simpler methods some protracted metab- olism experiments. I might incidentally mention the methods of adsorption and staining which hitherto have not been sufficiently considered. In the various food preparations (whose names need not be mentioned), it is quite unessential whether they contain a few per cent more or less of carbohydrate or nitrogen, a fact which is always especially emphasized in the advertisements; whereas it is quite important to know their swelling capacity, and whether this permits their complete and rapid utilization in the alimentary canal. Milk and Dairy Products. Milk, as a physiological excretion will be considered on pages 345, et seq., and there also much is said which pertains to its properties as a food material. Here we shall concern ourselves merely with the examination of milk. The present-day methods of milk examination are limited to certain characteristics which are especially easy to determine and, therefore, are easily simulated by adulterators. Pure food officials lay most stress on the water and the fat content. Sometimes, in addition, they de- termine the protein percentage, preservatives, and the possible ex- istence of disease organisms. Inasmuch as milk is by far the most valuable food-stuff, it is of the greatest importance not only to determine variations produced under the normal circumstances by adulteration, but also those occurring under normal conditions of production, change of fodder dependent upon change of season, natural and artificial fodder, boiling, pasteurization, etc. Accord- ingly, H. Zangger*^ and his pupils undertook to discover new methods; in them he regarded milk as a solution of colloids and electrolytes Of the colloid methods, Zangger and his pupil, KoBLER, have chosen the determination of surface tension, which proved to be one of the "most complicated but perhaps the most deUcate and flexible method." Among the various procedures, the bubble method, in which bubbles are allowed to form in the fluid, gave the most constant results. Normal milk gave quite constant figures. Inasmuch as by this method only such substances have an influence as are forced to the surface, these can make themselves evident in the minutest quantities. Adulteration with water is not easily detected by this method. Fermentation, on the contrary, causes great departures from the normal, which are explained by the development of fatty acids. The addition of alkalis also changes the surface tension. By the study of the viscosity, abnormal protein and fat content could be sho-s\Ti and likewise additions (adulterations) which in- fluenced the amount of swelling (especially alkaline additions) . The viscosity is also diminished by violent shaking, though milk regains 174 COLLOIDS IN BIOLOGY AND MEDICINE through quiet its original viscosity (to within 1 per cent) provided it has not been shaken long enough for curds to form. This observation was of great practical importance because milk suffers violent shaking during transportation. Experiments in which milk was carried by wagon, train and post more than three hundred (300) kilometers showed that there was no evident irreversible loss of viscosity. Dr. Grosser, according to a personal unpublished communication, has made a very noteworthy observation in the ultrafiltration of milk} It was shown that raw milk gave an ultrafiltrate much richer in lime than did boiled milk. In boiling, the calcium is bound to the milk colloid and remains with the latter on the ultra-filter. Thus, a simple means is furnished for distinguishing raw from cooked milk. Definite differences exist between human and cow's milk which offer a new basis for the difference these two kinds of milk exhibit in respect to their assimilibility (available food value). The classification of the milk colloids to which J. Alexander and J. G. M. BuLLOWA have drawn attention must be considered in future tests of milk (see p. 349). Since the water and the crystalloid content of milk are almost con- stant, many adulterants can be detected by the departure of the water and the milk content from the normal. For this purpose it is necessary to remove the fat and colloid constituents without chang- ing the content in water and salts. To determine the addition of water, J, Mai and S. Rothenfusser* coagulate the milk colloids with calcium chlorid and then measure the water content by refraction. Kurt Oppenheimer determines the milk sugar polarimetrically, after he has removed the milk colloids with colloidal ferric hydroxid. According to S. Rothenfusser, by treating milk with lead acetate in strong ammoniacal solution at 85° C, the milk sugar is adsorbed when the colloids are coagulated, while saccharose remains in solu- tion. According to Rothenfusser,* the smallest adulteration with foreign sugar (saccharated lime) may thus be detected. Among dairy products, condensed milk must be considered as of great importance. This is milk which is evaporated with the addition of 25 to 50 per cent cane sugar. All who are forced to use it, es- pecially colonists, know how ill it satisfies the demand for a milk substitute. One of the essential properties of colloids is that their condition is not reversible to the same extent as crystalloids. This may be, in addition to the destruction of certain flavoring substances, an important reason for the lessened value of condensed milk. The various dried milk products when stirred with cold or warm water ^ In a private communication, as yet unpublished. FOODS AND CONDIMENTS 175 give an incomplete emulsion and tiiere always «remains a sediment. The older the preparation the more incomplete is the solution. We here approach once more a phenomenon which was touched upon under "Aging of Colloids" (p. 74). J. G. M. Bullowa informs me that Just and Hatma ktcr have invented a process which avoids these disadvantages and which is already in use on a large scale.^ Cream is a fat emulsion which contains at least 10. per cent fat. Cream for whipping contains as much as 30 per cent. This emulsion has the property of building thick foam walls which possess con- siderable consistency. In order to simulate a high fat content, potato flour, gelatin or whipped white of egg are added as adulter- ants to cream deficient in fat. Calcium saccharate may also raise the viscosity. S. M. Babcock and H. L. Rüssel* recommended its addition to milk or cream wliich has become thin from being heated. The food industry has adopted this and, nowadays, calcium sac- charate solutions enber conmaerce under various names (grossin, etc.) as thickeners. According to Fr. Elsner, their effect is quite mar- velous. Their detection is easy by the method of Rothenfusser, described on page 174. An artificial cream may be obtained by emulsifying warm margarine with skim-milk and adding egg yolk. It is evident fiom what was said on pages 15 and 34, why an emulsion such as whipped cream or the like is so stiff, because we know how great a force is necessary to deform spheres of such small size. In milk and cream, the aqueous colloid solution is the dispersing medium and the fat is the dispersed phase; in the case of butter this relation is reversed.^ According to law, butter may not contain more than 16 per cent water, though it is possible to impregnate it with water to more than 30 per cent." According to Posnjak, the addition of alkalis and glucose increases, whereas increase of acidity diminishes the capacity of butter to absorb water. (W. Meijeringh.*) The kneading in of water is always reckoned to be an adulteration, because water is cheaper than butter. From the standpoint of the colloid chemist, it has always been a question whether the amount of water in butter or rather the content of skim- milk does not increase its digestibility and whether it is not the dispersion by means of the albumin or rather casern-containing aqueous solution which makes butter so much superior in digesti- bility to other fats of high melting point; and whether, if the above assumption should be proved correct, it would not be possible to find a legal way to permit butter to have a greater water (i.e., skim- milk) content. In the manufacture of margarine, skim-milk is 1 [Merrall and Soule of Rochester, N. Y., spray milk into heated air to dry it. Tr.] 2 [See work of Martin H. Fischer and G. F. L. Clowes. Tr.] 176 COLLOIDS IN BIOLOGY AND MEDICINE added to the fat in order, we are accustomed to assume, to give it the taste of butter. It is yet to be determined whether it is not just this addition, which gives the fat the dispersion characteristic of butter, and thereby its greater digestibility. The darkening of margarine in heating is undoubtedly to be attributed to the added skim-milk. By changing the colloidally dissolved albuminous substances of milk into the gel form, we obtain cheese. Coagulation can be brought about by means of rennet (sweet milk cheese) or through acidification (sour milk cheese). In cheese we have an emulsion of fat in a protein gel; whereas in skim or sour milk cheese (kümmel, Harz and hand- käse), the amount of fat is only the small amount afforded by skim- milk; it is quite high in the fatty cheeses (cream, Swiss, Camembert and Roquefort). A process of great interest, as yet uninvestigated from the colloid- chemical standpoint is the ripening of cheese. Through the action of bacteria there occur changes in the structure of the cheese which are specific for every variety and which cause the spotted appear- ance found in the various kinds of cheese. For cheese, chemical tests are limited to the determination of the water, fat, albumin and salt content and the possible adulterants. The most important, namely, the swelling capacity in the presence of the digestive ferments, is nowadays entirely ignored, although this would furnish the simplest method of deciding the important question of the digestibility of cheese. Honey should in the main be composed of sugar; it is neverthe- less frequently adulterated with glucose and dextrin. My opinion is that tests of the surface tension of dilute solutions would lead to the detection of such colloidal adulterants. Flour, Dough and Baking Products. The examination of flour, in addition to the microscopic histologic study, extends to its doughing and baking properties. These two questions belong entirely to the province of colloid chemistry. Art is in advance of science in this matter. There exist the most diverse methods for discovering the presence of foreign substances in flour and for distinguishing its various varieties by determining the temperature at which it becomes pasty. The baking capacity in particular, which is intimately bound up with the swelling capacity of the gluten, is applied colloid chemistry. The more glutenous the flour, the more water it binds (38 to 60 per cent) and the greater is its capacity to be kneaded. FOODS AND CONDIMENTS 177 The apparatus for this determination, especially that of Leo Liebermann, measures the expansion of the "doughed up" flour under the influence of heat. I must not omit to state here, that flour whose baking capacity- has suffered (for instance by over-heathig the gluten in grinding) may be restored by the addition of common salt, plaster of paris, water and alum. What has been said of flour applies also to prepared flours and in- fant foods. In the latter, in addition to the proper composition, ease of digestion and the property of preventing curdy coagulation of milk in the stomach must be considered. It should be determined whether a part of the difficult and complicated metabolism ex- periments could not be substituted by simple testing according to suitable colloid-chemical methods (swelling, etc.). In the investigation of dough, egg-dough and prepared products (bread, noodles and macaroni), there occurs a phenomenon which is very suggestive of a similar occurrence in the case of milk (see p. 345), namely, that one cannot recover the quantity of fat present in the original flour by means of ether extraction, and indeed, it would be interesting to determine how the adsorption of the fat occurs if there is any adsorption. In this connection we may consider that in the case of products made of egg-dough, we distinguish between free lecithin (extractible with ether) and hound lecithin (extractible with alcohol). Perhaps here, too, it is a question of adsorption. Next to milk, bread is our most important foodstuff. Bread making may be briefly mentioned here. Bread is prepared from flour, which, if it were consumed directly or made into a paste, would be badly digested because the flour grains possess only a small swelling capacity and the surface development of the entire mass is very small. The making of bread renders the individual parts easily accessible to the digestive juices. For this purpose, the dough (flour mixed with water) is caused to ferment by the addition of yeast or sour dough.' As a result, the starch grains swell, burst and take up water, a portion is converted into dextrin, part of which is still further broken down into sugar, alcohol and carbonic acid gas. By foam formation, the carbonic acid gas causes an enormous increase in the surface of the mass. Incidental fermentative processes give the gluten, the plant albumin, the power to swell up. This condition is completed and to a certain extent fixed by baking. The dextrinization of the starch is thus completed, the development of surface is increased b}^ the conversion of the water into steam and the expansion of the carbonic acid gas, the gluten is coagulated and further changes are stopped by the killing of the fermenting agents. 178 COLLOIDS IN BIOLOGY AND MEDICINE Ultimately we obtain a framework of coagulated gluten whose pores are filled with shattered starch grains. The crust, which does not swell much, acts as a protection both against the absorption of water and its loss from the interior. A good bread should contain from 35 to 45 per cent of water. Upon keeping, it loses about 1 per cent daily until the loss reaches 15 per cent. After that, the water content is dependent on the humidity of the atmosphere; this corresponds to the behavior of an elastic gel. It is interesting to note that the salt content plays a role in the condition of swelling, because unsalted bread dries much more readily than salted bread. There is a widespread error that stale bread has lost water and is dessicated. This is not true; the crumbling consistence of stale bread is due to a shifting of the water within the loaf; the starch grains transfer water to the albuminous framework. J. R. Katz studied this problem and found that bread kept fresh longer at 50° to 100° C. as well as below — 10° (best in a current of air), in other words, there is a balance of swelling in starch and gluten which corresponds to that of fresh bread. At from 0°-25° C. stale bread is the stable form. Staling is a particularly reversible process; dry rolls are made fresh by heating them. This is an old expedient frequently employed. The results of Katz' research on keeping bread fresh at low tem- peratures deserves the attention of the trade. The digestibility and available food value of bread depend par- ticularly upon its " dispersibility " and swelling capacity. A perfect wheat bread may be utilized to the extent of 94 per cent — a rye bread to 90 per cent. For this purpose, the flour used must be as fine as possible, otherwise the utilization is imperfect. It must also be properly swollen up; fresh bread which is too wet is digestible with difficulty. It packs together and the changes induced by the incorporation with the saliva and other digestive juices are differ- ent from those with old or dry bread. Most difficult to absorb are the proteins (55 to 85 per cent). The great heat (in the inner parts amounting to 110°), acting in the presence of a small quantity of water, produces a coagulation which greatly reduces their swelling capacity. The grain shortage during the war caused the attempt to make bread from potatoes. Potato flour yields a heavy, indigestible mass when heated in the way usual for ordinary flour. Different attempts were made to overcome this. According to A. Fornet, the Ex- perimental Station for the Utilization of Grain mixes in an unknown gluten substitute. Wilhelm Ostwald recommended blood or casein dissolved in ammonium carbonate as a substitute for gluten. FOODS AND CONDIMENTS 179 Walter Ostwald and A. Riedel made a porous starch bread by adding a starch paste to the starch dough before baking. A '' pseudo- coagulation" occurred during the baking, the unburst starch grains abstracting water from the burst ones. The internal friction of the paste becomes so great that the air bubbles cannot escape during the baking process, the dough does not fall but is fixed as a foam. [In discussing the physical chemistry of bread making, E. J. Cohn and L. J. Henderson (Science, Nov. 22, 1918, p. 501, et seq.) conckide that "the acidity of the dough, at the time of baking, seems to be the most important variable factor in bread making." Soluble serum protein is an acceptable physical substitute for gluten. Tr.] Beer. The fermentation industry so highly developed scientifically and technically has already paid attention to colloid chemistry and pro- duced a not inconsiderable literature (see F. Emslander *^) which in part, however, is not altogether free from dilettantism. It would take us too far afield, were we to consider the whole process of beer brewing ^ from the colloid-chemical viewpoint; we must restrict ourselves to the finished product. Beer is a fermented beverage with an alcohol content of from 2 to 5 per cent, some acetic acid and from 4 to 8 per cent extractives. The extractives consist in greatest part of carbohydrates (maltose, dextrin and gums), to a lesser extent of proteins (about 0.6 per cent), and in addition, salts, hop bitters, hop resin and several alkaloidal substances, besides small quantities of fermentation products such as glycerin, lactic acid and succinic acid. The persistent fine foam which a fresh beer should show is brought about by its colloidal content. It is a sign that the colloids have not j^et been broken down too far, and has at the same time the more important purpose of retaining the carbonic acid gas. In a solution supersaturated with gases, the formation of bubbles is either in- creased or diminished by the colloids present at the moment. We know further, from page 34, that a certain pressure is necessary to overcome the surface tension and burst the bubble, e.g., a soap bubble, so that the carbonic acid gas of beer is under a certain pressure beneath the foam. The condition rather than the amount of the foam-forming albu- mins is more important for the foam-keeping quality of beer. There are beers rich in albumin which remain foamless and beers poor in albumin which foam well. According to F. Emslander* it is mainly the soft hop resin, in addition to the acidity of the beer, which makes the albumins foam. 1 Rich. Emslander calls attention to an interesting relation between beer brewing and inactivation of ferments by shaking, see p. 189. Brewers have long known that shaking by trains, machines, etc., interferes with the fermentation and lagering of beer. 180 COLLOIDS IN BIOLOGY AND MEDICINE A perfect beer should be absolutely clear; turbid beers are unre- liable, but no objection can be raised to a dusty or net-like appearance. In the latter case the disperse phase consists of protein particles, dex- trins or precipitates of hop resin. Some yeast may also be suspended. Occasionally in very cold beers cloudiness develops which may be ascribed to precipitated albumins, which disappear when the beer is warmed. In the United States where iced drinks are in great de- mand, especial pains have been taken to master the difficulty. The alkalinity of the wash water, the carbonic acid, and the atmospheric oxygen during the brew, play an important role in the resistance of beer to cold. According to R. Emslander* the surest means is the addition of some pepsin. [Wallerstein has patented the addition of a proteolytic enzyme to beers, to prevent cold- cloudiness. Tr.] What is designated as "vollmundigkeit" or body in beer is caused by the colloid content. This property is almost identical with the viscosity, and is determined by the viscosimeter. If, for instance, the time required for water to run from a 50-c.c pipette = 1, and that of an equal quantity of beer = 1.43, we say that the "viscosity" is 1.43. It follows from what has been said on pages 152 and 153, that we may assume d priori, that the "vollmundigkeit" is largely dependent on the electrolytes, that is, on the content of acids and the kinds of salts. Even though the larger part of the salts is derived from barley, yet some are derived from the brewing water, and the hitherto partly unrecognized influence of the water may be attributed to this fact. E. MouFANG determined empirically the relation between optimum keeping quality, "full" and "palatable" taste, sediment and acidity. I refer to F. Emslander *^ for the colloid-chemical effect of the brewing water on lagered beer. Among the proteins, in addition to the gluten which flocculates out on boiling with acetic acid, peptone may be mentioned. H. Bech- HOLD * ^ was able to demonstrate only albumoses, upon examining a beer by ultrafiltration. Before generalizing, a large number of beers would have to be investigated in this way. It seems that E. Fouard has carried on such ultrafiltration experiments with starch solutions, worts and beer (cited by Emslander). W. H. van Laer has also made noteworthy experiments on the relationship between the ultra- filtrates of beer and musts, and their transparency. F. Emslander and H. Freundlich * have performed cataphoretic experiments and found that the colloids migrate to the cathode. In consideration of the acid content of beer, this finding is theoretically correct. R. Marc * has worked out a simple method for quantitatively determining beer colloids by means of the fluid interferometer. FOODS AND CONDIMENTS 181 A study should be made of the usefulness of several other colloid- chemical methods for the testing of beer, especially the determina- tion of surface tension, which might serve to distinguish the amount of various colloids contained, coagulation methods, etc. The mere suggestion should suffice. It must be mentioned in addition, that F. Emslander*^ has at- tributed to the "protective colloids" of beer a significance for the more easy adsorption of milk and other foodstuffs. Even earfier experiments, especially those of Ross van Lennep indicate that the presence of colloids and suspensions have an influence on the growth of microorganisms. S'hngen thoroughly investigated this matter in connection with alcoholic fermentation and obtained interesting results. He found that colloidal iron, albumin, silicic oxid and humic acid had no influence on alcoholic fermentation, but that, on the contrary, it was greatly hastened by turf, filter paper, blood charcoal and garden earth. He succeeded in proving the cause of this; the carbonic acid which is developed during alcoholic fermen- tation impedes fermentation and all substances which favor the dis- appearance of the carbonic acid favor fermentation. The action of the colloids mentioned is purely mechanical, somewhat like that of powdered glass, threads, wood chips or platinum shavings which hinder boiling. In the fermentation industry it is generally known that brewers grains and spun glass increase alcohoHc fermentation, and these phenomena have now been explained by Söhngen's investigation. S. Rothenfusser * has employed his colloid-adsorption method for detecting saccharose in the most diverse kinds of foods and condi- ments (wine, weissbier, cafö parfait, kilned malt, pastry, etc.). In practice naturally many other questions will appeal to food chemists. It might be determined whether the availability of vege- table protein, which on digestion is only from 60 to 70 per cent, could not be increased by a suitable method of preparation. Colloid- chemical methods must unquestionably be utilized in the investiga- tion of fruit juices, jellies and marmalades. We must remember that these are frequently mixed with glucose, which, when undeclared, should be regarded as an adulteration. Glucose contains in addition to dextrose, dextrin and unfermentable substances which may be determined by colloid-chemical analysis. Marmalades are adulter- ated with gelatin, agar-agar and isinglass. We trust that the mere mention of these facts may cause food chemists to give greater attention to colloid-chemical methods. CHAPTER XII. ENZYMES. For more detailed study we recommend the following books of reference: "The Nature of Enzyme Action," by W. M. Bayliss; "Allgem. Chemie der Enzyme," by H. Euler (J. F. Bergmann, Wiesbaden, 1910); "Die Fermente und ihre Wirkungen," by C. Oppenheimer (F. C. W. Vogel, Leipzig, 1913); [" Biochemical Catalysis in Life and Industry," by Jean Effront. Translated by Samuel Prescott. John Wiley & Sons, Inc., 1917. Tr.] List of the best known enzymes. Amylase hydrolyzes starches and glycogen into dextrin and maltose. Catalase decomposes peroxid of hydrogen. Chymosin is rennin. Diastase fluidifies starches and hydrolyzes them to maltose. Emulsin hydrolyzes glucosides. Erepsin hydrolyzes albumoses and peptones to amino-acids. Fibrin-ferment an hypothetical ferment which coagulates fibrin. Invertase hydrolyzes cane sugar. Lipase hydrolyzes fats into fatty acids and glycerin. Maltase cleaves glucosides. Oxidase an oxygen carrier. Pancreatin from the pancreatic juice is a mixture of several enzymes. Papain hydrolyzes albumin. Pepsin hydrolyzes albumin in acid solution. Ptyalin is the amylase of the saUva. Rennin coagulates milk. Steapsin is hpase. Trypsin hydrolyzes albumin in alkaHne solution. Tyrosinase oxidizes tyrosin and some of its derivatives. Zymase splits sugar into alcohol and CO2. To split complex molecules, chemists have to employ powerful re- agents, such as acids, alkahs, etc. They smash, as it were, the clockwork with a hammer and then pick out the undamaged particles. Just as a watchmaker employs for each screw a suitable tool or a specially made pliers, so nature has constructed delicate instruments for this purpose. Enzymes are such tools for the chemical break- ing down or building up of molecules. Albumin, carbohydrates and fats may all be split up by acids. For each purpose nature has a special enzyme, or even several; for the cleavage of albumin, pepsin and trypsin; for starches, diastase; for fats, lipase. We shall see that some enzymes are fashioned exactly for their use, so that the simile of Emil Fischer, which compares the enzyme 182 ENZYMES 183 to a key and the substance split up to a lock, is a very happy one. The simile can be extended still further, since the key may unlock thousands of similar locks and fails only when the key is worn out. Moreover, only very small quantities of enzymes are needed which are utihzed over and over again. This conception of enzymes corre- sponds with our present-day chemical conception of catalyzers. These latter are substances which either bring about or accelerate chemical reactions, without themselves figuring in the end products. For in- stance, platinum hastens the combination of O and SO2 into SO3, or the union of H2 and 0 into H2O. It is the nature of catalyzers that they split up compound sub- stances and build up the same substances from the cleavage products until a definite equilibrium is obtained. Thus, ricin, the enzyme of the castor bean, not only splits fat into glycerin and fatty acid but also unites glycerin and fatty acids into fats. The equihbrium is frequently one in which the synthetic action seems very subordinate. Thus, amylase under certain circumstances splits up 99 per cent of starch, yet it forms possibly but 1 per cent of starch from maltose. This, being a colloid, precipitates from the solution, and thus permits the formation of another 1 per cent of starch which gradually ap- pears as starch grains or glycogen and thus permits the further for- mation of starch. Hitherto, it has been impossible to obtain an enzyme in pure form and only by its stronger or weaker activity was it known whether a dilute or a concentrated preparation of enzyme existed. W. M. Bayliss rightly calls attention to the fact that, on account of their colloidal nature, enzymes always carry down by adsorption portions of the solutions from which they are obtained. It is, therefore, not surprising that albumin reactions are obtained from pepsin and tryp- sin and a carbohydrate reaction from amylase and invertase. In many cases it is possible to decide by means of diffusion whether mixtures are present (see R. 0. Heezog and Kasarnowski*). Ac- cording to L. RosENTHALER,* the presence of albuminous substances is of biological importance, protecting the enzyme from many de- structive influences, especially from H and OH ions. As the result of the constant presence of adsorbed impurities, we know almost nothing concerning the chemical nature of enzymes. However, we do know that all enzymes are colloids. S. Fränkel assumes that pure diastase is very greatly dispersed but as yet no sufficient evidence has been adduced; neither observa- tion in the ultramicroscope nor filtration through a 4 per cent ultra- filter is conclusive. To us it merely seems probable that such diastase is more highly dispersed than hemoglobin. 184 COLLOIDS IN BIOLOGY AND MEDICINE Even the colloidal state itself, i.e., great surface development, under certain circumstances, may be responsible for work similar to that per- formed by certain of the enzymes. For instance, G. Bredig catalyzed hydrogen peroxid by means of metal sols, particularly platinum sol, which he prepared by electric pulverization; that is, he obtained a result, the splitting off of oxygen, which in all appearances resembles that brought about by catalase, a ferment which occurs in blood, in milk, and in many plant and animal tissues. On this account G. Bredig called his metal sols "inorganic ferments," although with enzymes (or ferments), they share other properties to which we shall return. The action of enzymes is explained in part by their colloidal nature. In the organism they act chiefly on colloid substances (e.g., foods) with very extensively developed surfaces, so that under certain circumstances enzymes may be merely mechanically adsorbed. They consequently act upon the substrate in the greatest concentra- tion. It was shown by numerous adsorption experiments with indiffer- ent suspensions (charcoal, kaolin, cellulose) that enzymes have a strong tendency to concentrate on surfaces. It is possible to remove the rennin or pepsin (M. Jacoby*), and trypsin (G. Büchner and Klatte*) from solution by means of fibrin flakes or other coagulated albumin, or diastase by means of starch (H. van Laer). The reagents and sometimes also the products of the reaction are adsorbed by the colloidal enzymes. If the former accumulate, in accordance with recognized laws, the progress of the reaction is slowed. An example is the breaking down of hydrogen peroxid by catalase. The oxygen formed by the breaking down of H2O2 into H2O and O is ad- sorbed by catalase and the reaction is slowed (Waentig and Steche). Some enzymes, especially pepsin and papagotin, according to Rohongi, give reversible precipitates in salt-free, neutral solutions of different albumins on which they act. The inhibition of the action of an enzyme by a suspension or a colloid may be removed again under certain conditions by another indifferent colloid. If the activity of rennet has been destroyed by charcoal or normal serum so that the mixture no longer produces curdling of milk, the activity of the rennet may be restored by the addition of saponin. Somewhat different modifications are obtained by the addition of Cholesterin or by combinations of trypsin with charcoal, saponin or Cholesterin (Johnson, Blüm*). In this way, the numerous possibilities which result from the interaction of enzyme and antienzyme (q.v.) rest on a physical basis. The essential difference between an indigestible adsorbent and one which is dissolved by the enzyme is that the combination, e.g., be- ENZYMES 185 tween animal charcoal and trypsin, is primarily irreversible. The trypsin is fixed, water is unable to tear the trypsin from the char- coal though casein is able to do so (S. G. Hedin). It is seen, accord- ingly, that trypsin undergoes a change on the surface of charcoal similar to that undergone by dyes on fibers and if the process were to occur in the organism as it does on charcoal, the trypsin would be permanently withdrawn from the mixture. If, however, the sub- strate is digested and crystalloid cleavage products result, e.g., in the cleavage of fibrin by trypsin, the adsorption ceases of its own accord and the enzyme becomes free for further use, acting like a true catalyzer. This serves to explain the significance of the surfaces of the sub- strate on enzyme action. The action increases in speed in accord- ance with extent of surface per unit of weight. E. Abderhalden and Pettibone* demonstrated this in the digestion with pancreatic juice of albumen coagulated in different ways. In the case of enzymes their electrochemical nature is more im- portant than in the case of other colloids, and influences their ad- sorption. We frequently observe that if the proper H or OH ion concentra- tion is absent, an enzyme acts feebly on its substrate. Many neutral salts favor enzyme action; others inhibit it. For instance, pepsin acts strongly in acid solution only, trypsin in alkaline solution only. The investigations of L. Michaelis*^ and his co-workers show that the electric charge of different enzymes varies (see H. Iscov- Esco*^) and that, proportionately to the charge, they are unequally adsorbed by various substrates. We have previously mentioned that electropositive gels or suspen- sions, e.g., ferric oxid, completely adsorb electronegative solutions, e.g., serum albumin. An electronegatively-charged mastic or kaolin suspension completely attracts to itself serum albumin, only when it has become electropositive by acidification. These investigations gave the following results for a group of enzymes. In the table (see p. 186) X signifies a pronounced electric migration (to cathode or anode), i.e., complete adsorption; 0 signi- fies no migration or no adsorption; X — 0, 0 — X, respectively, more or less migration or adsorption. 186 COLLOIDS IN BIOLOGY AND MEDICINE Migration towards Adsorbents Ferment. Anode + Cathode + Iron oxid, clay, etc. Kaolin, mastic, arsenious sulphid, etc. neutral charcoal. Invertase: in neutral solution in acid, solution. . . X X 0- X 0 X X 0 X X X destroyed 0 0 X X -0 0 0 X 0 0 0 destroyed X X X X 0- X X X X X X X X -0 X X destroyed 0 0 0 0 X 0 X X X X X 0- X 0 X destroyed 0 0 destroyed X destroyed X X in alkaline solution Plant diastase: in neutral solution in acid solution X X X in alkaline solution Salivary diastase: in neutral solution in acid solution 0 X X in alkaline solution Trypsin: in neutral solution in acid solution X X -0 X in alkaline solution Pepsin: in neutral solution. in acid solution 0- X X X in alkaline solution Rennin (from pepsin) : in neutral solution in acid solution destroyed in alkaline solution Rennin (Grübler): in neutral solution in acid solution in alkaline solution We learn from this table that analysis by simple adsorption may replace the more difficult and complicated electric migration. It is very instructive for our knowledge of enzymes, that their action is strongest with the reaction which expresses their own charges as may be seen from the following table (from L. Michaelis) : Optimum Activity Occurs with an H-ion Concentration of Water 1 . 10-^ Invertase 2 . 10~^ Maltase 2.5.10-^ Trypsin 2. 10^ Erepsin 2 . 10-^ Pancreatic lipase 2 . 10~^ Pepsin 2. 10-2 ENZYMES 187 Electronegative (acid) "pepsin digests best in an acid; and ampho- teric (almost neutral) trypsin in an alkaline reaction. We may think of enzymes as being amphoteric substances, in some of which the positive charge predominates, in others, the negative; as a corollary of this, pepsin dissolves in alkaline solution; in other words, the pepsin is dissolved away from the substrate it would digest and is thus in- activated; the reverse of this holds true for trypsin. Salivary dia- stase seems entirely neutral, since saliva must fluidify as readily in acid as in alkaline reaction. In some cases we observe a relationship between the reaction of the substrate upon which a ferment is to act and the ferment; thus, pepsin-rennin has a pronounced basic character, casein an acid character, albumin in acid solution has a basic character, and as such combines with acid pepsin; in alkaline solution it has an acid character and so can unite with basic trypsin. Consequently, we find here phenomena which I pointed out in my experiments on the adsorption of dyestuffs, page 29. This difference in adsorbability is utilized in many cases for the 'purification of enzymes. Thus L. Michaelis removed albumin from mixtures of serum albumin and invertin by shaking them in acid solution with kaolin; the invertin remained without loss of strength in the albumin-free solution. E. Abderhalden and F. W. Strauch extracted pepsin from the stomach content of animals by means of elastin and then recovered it from the elastin by means of water. Depending upon the reaction, decided differences were found when enzymes were filtered through Chamberland filters. Accord- ing to Holderer most of those studied by him passed through the filter when they were neutral to Phenolphthalein but they were held back when neutral to methyl orange. Holderer attributes this principally to the effect of adsorption by the filter mass. For a number of enzymes the course of the reactions was studied and proved to be quite complicated. I refer to the investigations of platinum sol by G. Bredig and his pupils; of invertase and amylase by V. Henri; of lipase by M. Bodenstein and Dietz and of emulsin by P. Jacobson, which are described by H. Freundlich in his " Kapillar chemie." It is conclusively shown in the publication of W. S. Denham* from Bredig's Institute that among all the compli- cated factors, it is the surface concentration which is of the greatest importance for the acceleration of the reaction. In the case of gels, the greater the surface concentration becomes, or in other words, the more swollen the substrate is, the better is the opportunity offered an enzyme to enter the substrate. This obser- vation can be made over and over again in the cases of enzyme cleavage. E. Knoevenagel* offers convincing proof of this fact. 188 COLLOIDS IN BIOLOGY AND MEDICINE He says "The degree to which acetyl cellulose swells, runs parallel with its speed of saponification by aqueous alkalis, so that with greatly swollen acetyl celluloses the saponification by 1/2 n KOH is completed quantitatively in a few hours, at room temperature." We have not yet answered the question : What property is to be ascribed to the specific action of enzymes? We may regard this query as being answered by G. Bredig and Fajans* as far as the principle involved is concerned. They demonstrated that right and left campho- and bromo-campho-carbonic acid (which resemble one another like a picture and its reflection in a mirror) may be split into camphor and carbonic acid by bases acting as catalyzers. The speed with which these two opposites are broken down differs considerably if optically active bases (quinine, quinidine, nicotine and cinchonine) are employed, and may amount to 50 per cent. This is analogous to the specific action of enzymes upon chemically known substances with catalyzers which have definite chemical characteristics. It has been shown for most enzyme actions, with certain exceptions among the sugars, that of two optically active isomers both are attacked by a given enzyme, but one is always affected more quickly. We thus have a complete analogy for natural enzymes, but beyond this point the "lock and key" idea fails, since under no circumstances could "an asymmetric key fit the mirror image of its proper lock." But the analogy extends further. G. Bredig and Fiske effected asymmetric synthesis by a catalyzer of known composition. According to L. RosE-NTHALER, the enzyme action of emulsin accelerates the following reaction: C6H5.CHO+ HON = CeHs • CH(OH) • ON. Benzaldehyde + hydrocyanic, nitromandelic acid, acid G. Bredig. and Fiske replaced emulsin with quinme, but if they employed Chinidin as a catalyzer they obtained laevo rotary nitro- mandeHc acid in addition to the inactive product. We may summarize our present understanding of enzyme action thus: As a result of their colloidal properties, under favorable ex- ternal circumstances, enzyme and substrate are greatly concentrated at their interfaces, so that the course of the reaction is very much accelerated; the reaction between enzyme and substrate is purely chemical, conditioned by their mutual chemical constitution or con- figuration. [Ultramicroscopic observations suggest that possibly physical action is also involved. J. Alexander, Jour. Am. Chem. Soc, 1910, vol. 32, p. 680. Tr.] Enzymes, perhaps, exhibit the property of aging to a greater ex- tent than all other colloids. Some, e.g., trypsin, if dried, lose their ENZYMES 189 activity after a time; in solution they all deteriorate more or less rapidly. We do not know whether this depends upon purely me- chanical variations, or whether it is associated with a chemical change. For the former view we have the fact that many enzyme solutions may be inactivated by mere shaking; a rennet solution, for instance, need be violently shaken only two minutes in a test tube in order largely to deprive it of its capacity to coagulate milk. Even E. Abderhalden and M. Guggenheim* had observed that tyro- sinase, expressed yeast juice, and pancreatic juice had their activity partially inhibited by shaking them for 24 hours. A. O. Shaklee and S. J. Meltzer* found the same true for pepsin and M. M. Harlow and P. G. Stiles* for ptyalin. Quite independently in 1908, Signe and Sigval Schmidt-Nielsen* observed the inadivation of rennet by shaking, and subjected the phenomenon to a thorough study. It was deduced from this that inactivation by shaking is a surface phenomenon; the inactivation in- creases with the length of time and the violence of the shaking; the volume of air present, the concentration of the enzyme, and the tem- perature are all influencing factors. The enzyme becomes con- centrated in the foam and on the surface of the vessel employed. The foam is more active than the fluid and the procedure offers a possible method of concentrating enzymes. If a rennet solution that has been shaken is allowed to stand, it recovers some, but never all of its original activity; a portion remains irreversible. If saponin is added to a rennet solution, no inactivation results from shaking because saponin drives the rennet from the surface. Subsequently M. Jacoby and A. Schütze* published an analogous observation. They found that hemolytic complement (see p. 196) of guinea-pig serum was inactivated by shaking it at 37° C. Reac- tivation, in other words the reversibility of the process, depends on the duration of the shaking. At first, only a definite fraction of the complement is irreversibly inactivated by the shaking since it may be reactivated by "end piece" and also incompletely by "middle piece." When the shaking has been sufficiently prolonged the complement is irreversibly inactivated according to Ritz. The inactivation depends, according to P. Schmidt and Lieber, on the fact that the serum is made turbid by shaking, a foam is formed into which the globulin separates, and this globuHn adsorbs the complement. The reactivation by "end piece" results from the solution of the flocculated globulin thus liberating the complement (see p. 196). To what extent the action of the alkali (from the glass) assists in the inactivation has not been determined with certainty. On the contrary, it seems from the data, that only a portion of the 190 COLLOIDS IN BIOLOGY AND MEDICINE complement is inactivated, since it may be reactivated by addition of ''end piece" and of ''middle piece." Serum becomes turbid on shaking, and it is the author's opinion that this is evidently due to a coagulation of the albumin by shaking. In many investigations, especially in "immunity studies," it is cus- tomary to shake the test tubes, and I believe that some of the dis- agreements in the results of experiments by different investigators are due to a disregard of such surface phenomena. [R. Ottenberg does not consider this factor in his exhaustive study. Arch, of Int. Medicine, vol. xix, pp. 457-492. Tr.] The diffusion coefficient of several enzymes was measured ^ by R. 0. Herzog and H. Kasarnowski.* They are for Pepsin 0.062 (at 12° C.) Pepsin 0.066 (at 16° C.) Rennin 0.062 (at 16° C.) Invertin 0.032 (at 16* C.) Emulsin 0.033 (at 15.3° C.) From these figures the following molecular weights were cal- culated : Pepsin 13,000 Invertin 54,000 Emulsin 45,000 ] In studies on the filtration, ultrafiltration and diffusion of enzymes through membranes it must be determined beforehand, whether the filter adsorbs too strongly. Thus, e.g., a Chamberland filter per- mits no pepsin, trypsin, lipase or zymase to pass through, though the pores are of ample size. By choosing suitable membranes, these methods of separation have given valuable results. It has been possible by diffusion and ultrafiltration to separate a num- ber of enzymes, which were formerly regarded as individual, into two constituents having different properties. Thus, according to S. Fraenkel and M. Hamburg,* diastase prepared from malt may be divided into two enzymes. The one which diffuses changes starch into sugar, whereas the other merely fluidifies the starch. A. VON Lebedew* ultrafiltered expressed yeast juice and thus succeeded in demonstrating, that in fermenting sugar the disap- pearance of the sugar and the formation of carbonic acid are two distinct processes. In the course of such experiments it has been shown frequently, that the components are inactive individually and only exert their enzyme action in combination. The first observation of this kind was that of R. Magnus,* who dialyzed liver extract. The extract which originally split up fat thus became inactive; when Magnus * The figures are the mean of several determinations. ENZYMES 191 J- united residue and dialyzate, the mixture recovered its lypolytic properties. A. Harden and W. J. Young* made a similar observa- tion when they ultrafiltered expressed yeast juice. The filter res- idue lost its ability to cause fermentation but regained it when mixed with the filtrate. It is evident from this, that some enzymes consist of a colloid and a crystalloid constituent; the latter follow- ing the suggestion of G. Bertrand is called co-enzyme or co-ferment. In still other ways the co-enzyme shows crystalloid properties; unlike the colloid portion it is insensitive to boiling and frequently consists of a substance whose composition is well known. Thus, e.g., according to 0. von Fürth and J. Schütz,* sodium cholate and sodium glycocholate are co-enzymes of lipase; and according to BiERRY and V. Henri* the chlorin and bromin ions of alkaline salts are the co-enzymes for the action of pancreatic juice upon starches. In contradistinction to the co-enzymes, the anti-enzymes are usually colloids. Anti-enzymes are substances which interfere with the action of enzjmies. They are like the antitoxins which detoxi- cate toxins, and like the antitoxins, they occur to some extent in normal serum, or may be produced in it by the injection of enzymes. For instance, horse serum contains a large amount of anti-rennin which inhibits the coagulation of milk by rennin. By injecting the proper enzyme, anti-enzymes for lipase, emulsin, amylase, pepsin, papain and urease have been obtained. An exception to this is anti- trypsin which seems to be a crystalloid since it diffuses readily. It is the anti-enzyme which protects intestinal parasites from digestion by the pancreatic juices. According to S. G. Hedin,*^ the relationship between enzjrme and anti-enzyme is an adsorption, which probably results in a fixation. A certain similarity between enzyme and co-enzjntne is possessed by pro-enzyme and its activator. Most enzymes are formed in an inactive state called the pro-enzyme, pro-ferment, or zymogen, which becomes active only after the addition of some crystalloid, usually a simple substance. The pro-enzyme of pepsin may be extracted from the gastric mucous membrane, but cannot, digest albumin; only after the addition of very dilute acids does it become pepsin and acquire its ability to digest. The trypsin of pancreatic juice is excreted into the duodenum as an inactive pro-enzyme; it is activated by calcium salts. This is of the greatest biological significance, since otherwise, secreting glands would not be safe from their own secretions. According to E. Pribram,* the formation of pro-enzymes occurs in this manner; the protoplasm of the glandular cells retains a cer- tain portion of the food by adsorption. Acids, calcium salts, etc., arrest the adsorption, and the active ferment becomes free. In 192 COLLOIDS IN BIOLOGY AND MEDICINE support of this view, E. Pribram and E. Stein injected through a tube into the stomach of one rabbit an active solution of rennin, and into the other, one mactivated by boihng. After four hours the rabbits were killed and the amounts of pro-ferment contained in their gastric mucous membranes were measured. The gastric mucous mem- brane of the rabbit treated with active rennin contained much more pro-enzyme than the other. From this, the authors deduced that the colloidal gastric mucous membrane adsorbed the rennin and changed it into pro-enzyme. S. G. Hedin* also views zymogen as the com- bination of an enzyme (rennet) with an inhibiting substance. In the condition studied especially by him, rennet is freed by hydrochloric acid, the inhibiting substance is let free by ammonia, with the destruction of the rennet. CHAPTER XIII. IMMUNITY REACTIONS. That the organism is overwhelmed by a large dose of poison but recovers from a small one should not particularly surprise us. Ever smce the recognition of the nature of infectious diseases, it must have amazed biologists that every infected organism did not succumb to the sUghtest infection. Microorganisms multiply indefinitely, and, theoretically, it is only a question of hours before the number present shall be overwhelming whether the infection is with a large or a small dose. Were this assumption, to which we might be led from the ob- servation of culture media, correct, no living thing, plant or animal, could exist. There must be inherent forces in the living organism which protect it agaihst pathogenic germs, which make it immune to such injuries, and which are called, accordingly, immune bodies (immune substances). L. Pasteur was the pioneer in the systematic study of immunity. He produced experimental proof that immunity might be artificially produced by previous treatment with attenuated infective agents (chicken cholera) just as had been done previously, in the case of vaccination against smallpox. These investigations received a mighty impulse when Robert Koch succeeded in growing disease germs in pure culture. The doctrines of immunity and predisposition were developed into a special branch of science which at present holds the chief interest of scientific medicine. It was recognized that the body could overcome its invaders in various ways: substances occur which make bacteria harmless by dissolving them, bacteriolysins (acting against vibrios, e.g., of cholera, and against typhoid), and others which clump them together and precipitate them, the agglutinins (against typhoid, paratjq^hoid, dysentery, etc.). In other cases, the protection is directed prin- cipally against the poisons, toxins, which the organized germs de- velop (diphtheria toxin, tetanotoxin, etc.). The organism possesses a peculiar protective mechanism in the leucocytes, which take up and digest the bacteria and cocci, devouring them like free living amebse in search of food. This phenomenon, which was recognized and studied chiefly by E. Metschnikoff, is called phagocytosis. 193 194 COLLOIDS IN BIOLOGY AND MEDICINE The bacteria are previously prepared for phagocytosis by certain Immune substances in the serum called opsonins. The study of these phenomena was very much simplified, when it became possible to transfer many of them from the living organism to test tubes. They were thus freed from disturbing epiphenomena and made susceptible to quantitative investigations. By these methods of study, we have learned a number of properties of the blood and cells, which have no direct influence on the natural pro- tection of the organism against the attacks of microorganisms, or which may be regarded merely as epiphenomena. They lead to the knowledge that the weapons of the organism against disease germs are not teleologically forged for this sole purpose, but that they are the product of a universal biologic law, according to which the organism produces antisubstances against all kinds of substances foreign to the species (art-fremde). In accordance with their historic recognition, and the method of their investigation, it is customary to class them with immunity phe- nomena: I am referring to the substances which dissolve and floccu- late blood corpuscles, the hemolysins and hemagglutinins and the albumin-precipitating substances, the precipitins: and finally the Wassermann reaction in syphilis, and anaphylaxis. . If the sera of two animals, e.g., cattle serum and rabbit serum, are mixed, the solution remains clear. If an animal, e.g., a rabbit, is in- jected with the serum from a different species of animal, e.g., cattle serum, substances are formed in the rabbit, precipitins.'^ If we then mix the serum of such an animal, ''cattle-rabbit," with ox serum, a precipitate forms. Agglutinins and hemolysins develop in a similar way. If a rabbit has cattle blood corpuscles injected into its veins, substances develop in the rabbit serum which agglutinate and dissolve the cattle blood corpuscles. Hemolysin consists of two substances, one heat resisting (thermostable) and specific, the amboceptor, and another, heat sensitive (thermolabile, destroyed at 55° C.) and non- specific, the complement. Only the amboceptor develops as a result of injecting the red blood corpuscles, the complement is always present in every serum. However, both are required for hemolysis. We have now explained the formation of precipitins for cattle serum or blood corpuscles in rabbits, but the principle is of general applica- tion for the injection of blood into a different species of animals. To ^ The so-called "precipitin reaction" is of great medico-legal importance. It serves for the differentiation oi human and animal blood, for which a small drop suffices. It is also employed in detecting adulterations (horse-meat in sausage, etc.). In the study of phylogenesis it is a valuable aid particularly in teaching the natural relationships of animals. IMMUNITY REACTIONS 195 make the application general, if an animal is injected with sub- stances foreign to its species {antigen), e.g., albumin, animal cells, bacteria, toxin, antibodies or immune substances (precipitins, hemol- ysins, agglutinins, antitoxins) are formed in the injected animal. Binding of antigens (toxin, bacteria, etc.) by the immune sub- stances (antitoxin, bacteriolysin, etc.) results from combination or a sort of neutralization which may be compared to the neutralization of an acid by a base. P. Ehrlich* was the first to study this neu- tralization quantitatively and showed in the case of diphtheria toxin and its antitoxin, that the saturation did not occur as in the case of a strong acid, e.g., HCl, and a strong base, e.g., KOH, but that the diphtheria toxin must consist of a mixture of more or less acid toxins. We reach this conclusion, not only from the course of the saturation curve, but also from a study of the different poisonous actions pos- sessed by the individual saturation fractions. Though, e.g., the larg- est part of the diphtheria toxin has an acute toxic action, there is a particular fraction, the toxon, which, after two or three weeks, pro- duces paralysis of the extremities that are quite foreign to the toxin. In different cases various indicators are used as a sign of the union between antigen and immune substances. In the case of toxin-antitoxin ^ we are reduced to the biological proof by animal experiments; the re- duction in the toxicity of the mixtures is determined from the toxic action remaining in them. In the case of hemolysins, the ability to dissolve red blood cells more or less completely is used as a sign. Precipitins are recognized by testing the antigen against various dilutions and determining the greatest dilution at which turbidity can still be recognized. If, e.g., a ra;bbit has been injected with goat serumi a substance develops in the rabbit which precipi- tates goat serum. If we add to goat-rabbit serum which has been placed in a row of test tubes goat serum in a dilution 1/100, 1/1000, 1/10000, etc., we shall find a dilution at which merely tur- bidity occurs. In a similar manner agglutinins are tested (in this instance the immune serum is diluted). Since the nomenclature is not uniform, a table of the terms in com- mon use is given here. Agglutinins change bacteria so that they may even be precipitated by alkali salts (see antigen). Amboceptor, see hemolysin. Antigens, foreign substances (bacteria, proteins, toxins, etc.) against which specific antisubstances (antibodies) are developed by an animal injected with them (agglutinins, precipitins, antitoxins, etc.). Antibodies, immune bodies. Antitoxin, specific antibodies which neutrahze toxins. End piece, see hemolysin. ^ [Jerome Alexander observed in the ultramicroscope the mutual coagulation of diphtheria toxin and antitoxin and tetanus toxin and antitoxin; diphtheria toxin was not precipitated by tetanus antitoxin. Tr.] 196 COLLOIDS IN BIOLOGY AND MEDICINE Hemolysins dissolve red blood corpuscles. Two substances are usually required for hemolysis. One is specific, the real antibody, and is called amboceptor. The other occurs in every serum and is complement. Complement consists of two parts, one of which, the middle piece, is precipitated with the globulin; the end piece remains with the albumin of the serum. Only when both are united does complement act. According to H. Sachs, Omokvkow and Ritz there is an addi- tional " third component," quite heat stable. According to P. Schmidt comple- ment is a single substance of which a portion is adsorbed by globuhn when it is precipitated. Complement, see hemolysin. Lysins cause solution. Bacteriolysins dissolve bacteria, hemolysins dissolve red blood corpuscles. Precipitins flocculate albumin. Toxins, poisons which produce antitoxins when injected. The Nature of Antigens and Immune Bodies. The substances involved in immunity reactions are all dissolved or suspended colloids. There is, therefore, a particular reason for studying these questions from the standpoint of colloid investigation.^ So far it has been impossible to produce immune bodies by means of a crystalloid; a foreign colloid (antigen) has always been required. The proof of the colloid character of antigens and immune bodies has been demonstrated in numerous cases. Upon dialysis, they do not pass through a dialyzing membrane; the diffusibility of diph- theria toxin and tetanolysin and their antitoxin are indicative of a particle magnitude of the same order as hemoglobin (Sv. Arrhenius). Ultrafiltration of diphtheria toxin, toxon and antitoxin and anti-rennin gave similar results (H, Bechhold). The hemolytic complement of guinea-pig serum is inactivated by shaking with the formation of a precipitate (M. Jacob y and A. Schütze). This indicates a concen- tration at the boundary of fluid/air, as in the case of albumin and other colloids. The observation of W. Biltz, H. Much and C. SiEBERT that a bactericidal horse serum loses its bactericidal activity upon shaking is to be ascribed to a similar phenomenon.^ ^ We may mention the following papers which treat Immunity with particu- lar reference to the standpoint of coUoid chemistry: K. Landsteiner, Die Theorien der Antikörperbildung, Wiener Klin. Wochen- schr. 22, Nr. 47 (1909). Idem., Kolloide u. Lipoide in d. Immunitätslehre im Handbuch d. Pathogenen microorganism von Kolle u. Wassermann, Bd. II (1913). O. PoRGES, im Handb. d. Technik u. Methodik d. Immunitätsforschimg, Bd. II, Lief. 2 (Jena, 1909). H. Zangger, Viertel] ahrsschr. d. Naturf.-Ges. in Zurich, 1908, 408-455. 2 It might be claimed that these substances, which it is impossible to prepare in pure form, are not in themselves colloids, but that they are adsorbed by the proteins simultaneously present in the solution and thus simulate -what coUoid character they exhibit. For the correctness of this view no evidence tas hitherto been presented. IMMUNITY REACTIONS 197 It is held by one small group of investigators that antigens are hpoids or lipoid-albumin compounds. Since the part taken by lipoids in many immunity reactions is not definitely settled, it is impossible as yet to determine the general correctness of this view. At any rate it has not as yet been possible to immunize with the lipoids chemically known. Since our knowledge concerning the chemical composition of normal proteins is still meager, what we know about the proteins of immune bodies cannot be more ample. According to Kirschbaum dysentery toxin is acid. By ultrafiltration, he prepared water insolu- ble acid dysentery-toxin which was nearly atoxic, though the salt obtained by dissolving the acid in an alkaline carbonate possessed the poisonous properties of the toxin. The experiments of Fr. Obermayer and E. P. Pick* indicate that the aromatic nucleus in the antigen is of great importance for the development and character of the antibodies. Antibodies are universally regarded as albuminous. Before we discuss details, let us indicate a great misunderstanding which at the time gave rise to heated discussions and clouded the issues. The quantitative relations in which the substance in ques- tion enters into reaction (toxins with their antitoxins, bacteria with their agglutinins, etc.) have great similarity to adsorption curves (W. BiLTz) and to the neutralization curves of certain weak acids and bases (Sv. Arrhenius). These investigators laid great stress on this fact and believed that they had thus discovered the nature and the course of the immunity reactions in question. P. Ehrlich raised the weighty objection, that the reaction is specific, and that the poisons are very complex: diphtheria toxin is detoxicated only by diphtheria antitoxin; typhoid bacilli are precipitated only by typhoid agglutinin. There is no doubt that these specific proc- esses cannot be explained by what we call colloid-chemical reactions (see H. Bechhold*^) . We must conceive of the process as occurring in two stages, and we must emphasize that this sharp distinction does not obtain in every case. First Stage: The two colloids, toxin and antitoxin, bacterium and agglutinin, unite in accordance with the laws governing other colloids, e.g., fiber and dye, and the specific substances react on one another and it is still an open question whether we must represent these reactions as chemical or catalytic. Second Stage. The colloidal product of the reaction shows physical properties which distinguishes it from the reacting substances, e.g., it precipitates. We cannot enter here into the question of specific combination. 198 COLLOIDS IN BIOLOGY AND MEDICINE A. The Distribution of Immune Substances Between Suspensions and Solvent. Bacteria and blood corpuscles form suspensions which to a greater or less extent are able to attract to themselves immune substances. It is fortunate for the study of these phenomena that many experi- ments have been preformed upon the adsorption of inorganic sus- pensions (kaolin, charcoal, ferric hydroxid gel, etc.) from solutions of known composition. For comparison many investigators have per- formed appropriate experiments on toxins and immune substances. Adsorption by Means of Inorganic Suspensions and Hydrogels. A sign of adsorption is a strong withdrawal of a dissolved substance from dilute solutions and a relatively smaller withdrawal from such as are more concentrated. This requires extensive quantitative inves- tigation with solutions of different concentration. Unfortunately, there are but few such experiments published. It may, however, be deduced from the results of some of these experiments that an ad- sorption curve is actually involved. Even W. Roux and Yersin* found that calcium phosphate, aluminum hydroxid and bone black removed some poison from a solution of diphtheria toxin, but that the solution was never entirely detoxicated. W. Biltz, H. Much and C. Siebert* shook gels of iron oxid, chromium oxid and zirconium oxid, among others, ^s^ith tetanus and diphtheria toxin, tetanolysin and a bactericidal horse serum. They determined a diminution in the activity of the respec- tive solutions, and that for the same quantity of hydrogel, the diminution by activity was frequently more marked for dilute than for concentrated solutions. Occasionally complete fixation or destruc- tion occurred, e.g., in the case of typhoid agglutinin. H. Bechhold** found that arachnolysin and staphylolysin were never completely re- moved from solution by formol-gelatin or cellulose. K. Land- STEiNER and his pupils shook tetanus toxin with kaolin, protagon, Cholesterin, palmitic acid, stearic acid and lecithin; a poisonous re- siduum was always discovered in the solution. Since complement may be removed from a solution by various sus- pensions (for literature see H. Sachs), a mechanical adsorption is probable. Specific Adsorption. Glancing at the entire literature on this question, we are con- fronted with the great difference in adsorption capacity of the ad- sorbents as well as of the adsorbed substances. Although tetanus toxin is well adsorbed by kaolin, protagon, Cholesterin, palmitic acid, IMMUNITY REACTIONS 199 stearic acid, and lecithin, only very little of it is taken up by cetyl alcohol, casein, coagulated serum albumin, and starches (K. Land- STEiNER and A. Botteri*) . Arachnolysin is adsorbed more strongly by glacial acetic acid collodion than by formol-gelatin; glacial acetic acid collodion adsorbs rennin very strongly, but adsorbs practically none of a serum containing anti-rennin (H. Bechhold*^). Silicic acid and barium sulphate fix complement which, however, is also fixed by kaolin to a lesser extent (E. Hauler*). In view of these specific influences, L. Jacque and E. Zunz* un- dertook extensive experiments upon the adsorption of antigens and antibodies by inorganic suspensions. They concluded, as had pre- viously been shown by E. Zunz* that differences in surface tension were not alone determinative for adsorption. They found, e.g., that bone black strongly adsorbed diphtheria toxin as well as antitoxin, though neither was adsorbed by wood charcoal, diatomaceous earth, talc, kaolin or clay. Nevertheless kaolin and clay adsorb tetanolysin. Bone black, a good adsorbent for diphtheria antitoxin, does not adsorb the antitoxin of tetanolysin or cobra hemolysin. Reversibility. A purely mechanical adsorption demands that the process be completely reversible. This occurs in the case of the slightest adsorptions of immune bodies by unorganized suspensions. W. BiLTZ, H. Much and C. Siebert* have already called attention to the fact that the adsorption of their antigens by hydrogels was only slightly reversible. Only to this extent was J. Bordet's* com- parison of immune reactions to the dyeing of fiber with dyes appro- priate. This irreversibility has its analogies in the adsorption of numerous other known substances in which we assume that secondary changes occur as a result of the concentration at the surface; some of these changes are chemical, e.g., the adsorption of crystal violet and of rennin by bone black. Of great interest are the observations of L. Jacque' and E. Zunz* illustrating the competing action of several adsorbents for a single substance. They found that the adsorption of diphtheria toxin by bone black was reversible in the body but irreversible in vitro [probably because of protective substances. Tr.]. The adsorption of diphtheria antitoxin is, on the contrary, irreversible in the body and reversible in vitro. Serum albumin may prevent the adsorption of diphtheria toxin and antitoxin by bone black. Adsorption by Organized Suspensions. If agglutinin is added to bacteria, or hemolysin to blood corpuscles with the same quantity of the suspension, proportionately more ag- glutinin or hemolysin will be combined from a dilute solution than 200 COLLOIDS IN BIOLOGY AND MEDICINE from a solution that is concentrated. Eisenberg and Volk* dem- onstrated this for typhoid bacilU and cholera vibrios and Sv. Arrhenius* and his co-workers for hemolysins (see also G. Dreyer, J. Sholto and C. Douglas*). This is illustrated by a table (after Eisenberg and Volk) showing the combination of agglutinin with a uniform quantity of typhoid bacilli and increasingly concentrated agglutinin solutions. Agglutinin fixed by the Agglutinin free in the bacteria. solution. 2 0 20 0 40 0 i 180 20 340 60 1,500 500 6,500 3500 11,000 9000 The course is entirely that of an adsorption curve. Cases also occur, according to the investigations of G. Dreyer, J. Sholto and C. Douglas* in which, after exceeding a certain maxi- mum, less and less agglutinin is taken up by the bacteria, in spite of greater concentration of the agglutinin, typical "abnormal adsorp- tion." It may be mentioned, moreover, concentrated salt solutions inter- fere with the fixation of agglutinin (until now this had been demonstrated only for blood corpuscles). (K. Landsteiner and St. Weleck.) Analogous to this is an observation of W. Biltz,*^ according to which the addition of salt interferes with the adsorp- tion of proteins by inorganic colloids.^ As has been stated elsewhere, hemolysis with immune sera occurs through the interaction of two components; amboceptor is bound by the blood corpuscles and it causes the fixation of the complement which accomplishes the hemolysis. This is obviously very similar to what occurs with mordant and dye; the dye is fixed to the fiber by means of the mordant. K. Landsteiner and N. Jagic* have to a certain extent devised a model for the process; as amboceptor they use silicic acid hydrosol, as complement active serum or lecithin. Sihcic acid precipitates with blood corpuscles as well as with leci- thin; it thus links together blood corpuscles and lecithin, concentrating 1 In my opinion it is chiefly globulin which has been strongly adsorbed in these experiments, since the final portion of globuUn is separated very slowly from dialyzed serum. IMMUNITY REACTIONS 201 ./ the lecithin on the surface of the blood corpuscles. It is very prolj- able, as assumed by the investigators mentioned, that lecithin in this instance acts as a solvent for the lipoid membrane of the blood cor- puscles. Numerous similar models in which complement was re- placed by lipoid were subsequently devised. Reversibility. The combination of agglutinin with bacteria and red blood corpuscles is partially reversible; it may be partially removed at a higher temperature by washing with physiological salt solution (K. Landsteiner, Eisenberg and Volk). This is deduced from an experiment of J. Jogs.* He mixed typhoid bacilli bearing agglutinin with untreated typhoid bacilli, and all the bacilli became agglutinated. There must have been a with- drawal of agglutinin from the typhoid bacilli which had been treated. In a similar way, J. Morgenroth* ^ demonstrated that the combi- nation of amboceptor and red blood corpuscle is partially reversible. Reversibility within the organism, where numerous varieties of cells occur, is of great practical importance. B. The Distribution of Immune Substances Between Dis- solved Colloids and Solvent. The colloid-chemical theories regarding the combination of toxin and antitoxin are tacitly based upon the assumption that toxin and antitoxin behave like a suspension or a hydrogel; they premise that surfaces occur, upon which, for instance, the toxin may concentrate in accordance with the laws of adsorption. Theoretical basis for this assumption is lacking. Very little is known concerning the fixation of crystalloids and of hydrosols by hydrosols, when no precipitate occurs. There are, at present, two methods of attacking the problem. By means of ultrafiltration, H. Bechhold** has shown that the com- bination of methylene-blue with serum albumin satisfies the conditions of an adsorption (see p. 25). L. Michaelis and P. Rona have used the osmotic compensation method in order to determine the kind of combination in which sugar, Ca, etc., are fixed in the blood. Both methods are recent and had not previously been utilized in the solu- tion of this problem.» On this account I consider it unprofitable to dis- cuss at present the manner in which toxin and antitoxin are combined. It may be mentioned that the toxin-antitoxin combination is in- completely reversible in part and in part irreversible. P. Ehrlich and his pupils demonstrated by numerous biological investigations, that the combination between diphtheria toxin and its antitoxin rapidly became irreversible. The relations between precipitin and precipitable substance is some- 202 COLLOIDS IN BIOLOGY AND MEDICINE what more obvious. Inasmuch as the two dissolved colloids yield a precipitate when mixed in suitable proportion, we can form a judgment concerning the proportionate quantities that combine. However, this has regard for the composition after precipitation. Ac- cording to E. VON Düngern* the precipitate binds much more precip- itin than is required to cause complete precipitation. It is still a question whether this combination existed in the solution. C. Precipitation of Dissolved Colloids and Organized Suspensions, Serum containing precipitin, for instance, goat-rabbit serum, gives a precipitate with its antigen (goat serum). The serum is precipitated by the precipitin just as it would be hy an inorganic hydrosol or an acid protein (histone), U. Friedemann and H. Frieden- thal* (see p. 157). The precipitation occurs best in the presence of an optimum mass proportion between precipitin and precipitable substance; excess of precipitable substance interferes with the pre- cipitation. A precipitation, according to M. Neisser,* occurs also in salt-free solution (which contains no globulin) but the precipi- tation zone differs from that in a solution containing salt. Though the mutual precipitation of two amphoteric colloids de- pends on the hydrogen ion concentration (see p. 147), specific precip- itations (this applies to precipitins and agglutinins) are largely in- dependent of it (L. Michaelis and Davidsohn). The electric charge of the components plays a very subordinate part in these precipitations. The plant toxins, ricin (from the seeds of varieties of castor bean) and abrin (from jequirity seeds) have a similar precipitating effect upon albumin. The conditions in the case of organized bacterial albumin are similar to (but not identical with) those of serum albumin. If an animal (e.g., a rabbit) is injected with bacteria (for instance, typhoid bacilli) there develops in its blood an agglutinin which causes typhoid bacilh in a test tube to precipitate.^ Agglutinin forms a compound, with the (actual) albuminous capsule of the bacteria, so that these behave as though they were changed from a hydrophile to a hydrophobe suspension. Precipitation occurs only in water con- taining salt.2 Though a suspension of bacteria is unchanged by ^ This phenomenon was first observed by Gruber and Durham. Widal was the first to use it for diagnosis, and since, as the Gruber-Widal Reaction, it is em- ployed in the diagnosis of typhoid, paratyphoid, dysentery, etc. ^ According to U. Friedemann it is possible to obtain agglutination in a salt- free solution, though this has nothing in common with specific agglutination. The resemblance to the precipitins is, in this respect, only a superficial one. IMMUNITY REACTIONS 203 dilute alkali salts, these ^alts cause a precipitation of agglutinated bacteria as they would a suspension of kaolin or mastic. This was demonstrated by H. Bechhold*^ as well as M. Neisser and U. Fribdemann*, and practically confirmed by B. H. Buxton, P. Shaffer and 0. Teague*, as well as by B. H. Buxton and A. H. Rahe*. Capsulated bacteria possess, in their mucous capsules, a natural protective colloid and accordingly, they are not agglutinated by immune serum even in suspensions containing salt. 0. Forges showed that if the mucous capsules were removed by gently heating them with dilute hydrochloric acid, even capsulated bacteria were agglutinated by immune sera. As in the case of the precipitins, a certain quantitative rela- tionship between bacteria and agglutinating serum is required for precipitation. But in this instance, there are certain irregularities as regards native and heated sera. This phenomenon also has its analogue in the precipitation of unorganized suspensions in the presence of protective colloid. Table I (see below) illustrates the agglutination of bacteria by diluted immune serum. Table II illustrates the precipitation of a mastic suspension by AI2 (804)3 in the presence of leech extract as a protective colloid. These "irregular series" (see p. 84) frequently occur in the pre- cipitation of suspensions by ferric chlorid, aluminium chlorid and certain dyes. They are explained by the fact that the hydrolytically split iron oxid hydrosol, etc., functions as a ''protective colloid," and in certain proportions interferes with the precipitation. If still an- other protective colloid (gelatin, leech extract, etc.) is added, the circumstances are still further complicated as shown by Table II. TABLE I. TABLE II. Agglutination of typhoid bacilli with very dilute immune serum. (After Eisenberg and Volk.) Precipitation of mastic suspension by 0.0002 c.c. J Al2(S04)3 m the presence of leech extract as pro- tective colloid. (After M. Neis- ser and U. Friedemann, and H. Bechhold.) Dilution of the serum. Appearance after 2 hours. Dilution of leech extract. Appearance after 24 hours. 1/100 1/1000 1/25000 1/30000 1/45000 No a.gglutination Almost complete agglu- tination Trace Heavy flocks No agglutination c.c. 0.1 0.03 0.01 0.003 0.001 No precipitation Almost complete pre- cipitation No precipitation No precipitation Complete precipitation 204 COLLOIDS IN BIOLOGY AND MEDICINE Since bacteria, like other proteins, may be precipitated by acids without previous treatment with agglutinating serum, L. Michae- lis and Beniasch tested different groups of bacteria (typhoid, para- typhoid, colon, etc.) to determine whether the result depended on the H ion concentration. They found that different concentrations of H ions were necessary for the precipitation of different groups of bacteria. The discoverers based on this fact, a method for distin- guishing certain bacterial groups. It is still a question whether the procedure is practical either alone or in combination with agglutinat- ing serum (see S. Galitzer). Blood corpuscles, resembling bacteria, behave like a hydrophile suspension whose surface is so changed by various agglutinins that they flock out. I am inclined to believe that a true glueing together occurs more frequently with blood corpuscles than with bacteria. From our previous experiments, we see that colloids and suspen- sions are precipitated not only by electrolytes but also hy colloids of opposite charge. It must therefore be possible to agglutinate or- ganized suspensions as bacteria and blood corpuscles by suitable hydrosols. Experiments with hydrosols of iron, zirconium, thorium oxid and silicic acid ^ (W. Biltz, H. Much and C. Siebert*, also K. Landsteiner and N. Jagic*, Girard-Mangin and V. Henri*) con- firm this assumption and show that as with other colloid precipi- tations in salt solutions, an optimum proportion between the two colloids must exist, or no precipitation will occur. It should be emphasized that the combination of bacteria or blood corpuscles and inorganic hydrosols is irreversible (in contradistinction to the agglutinin combination). Blood corpuscles differ very greatly in one respect from bacteria. Though the latter migrate to the anode showing their negative charge, blood corpuscles are more amphoteric. As a result of this, they are precipitated by negative hydrosols (arsenic tri-sulphid, silicic acid, etc.) L. Hirshfeld*^. A very important observation is that of L. Hirshfeld, that the agglutination of blood corpuscles of different animals by zinc nitrate follows the same order of precipita- bility as their agglutination by agglutinating sera and abrin. Ricin and abrin also agglutinate blood corpuscles, obviously, be- cause they precipitate their albumin. From our entire exposition, it is evident, that the adsorption of the agglutinating substance and the agglutination are two separate proc- esses which, in their principle, have nothing in common. The agglu- tinin changes the bacteria or erythrocytes, making them agglutinable; 1 Colloidal silicic acid agglutinates in much greater dilution than the crys- taUoidal. IMMUNITY REACTIONS 205 the electrolyte, which itself is not adsorbed, agglutinates or flocks them out.i It is therefore obvious that an electrolyte which changes the cells may, nevertheless, without being adsorbed, agglutinate them. According to J. Dunin-Borkowski*, red blood corpuscles are agglu- tinated by FeCls, though it is not combined with them. Electric Charge, H and OH Ions. Numerous attempts have been made to determine the electric charge of antigens and immune substances by cataphoresis (K. Landsteiner and Wo. Pauli*, C. N. Field and 0. Teague*. It is so small, however, that in my opinion it cannot be definitely rec- ognized since traces of H or OH ions may cause a reversal of charge (see Bechhold*^*'). This also applies to the results of exhaustive adsorption experiments with adsorbents of opposite electric charge, namely, the experiments carried out by Edgar Zunz with electro-osmotically purified sihcic acid, aluminium hydroxid, kaolin, diatomaceous earth, talc, and clay upon toxins and antitoxins. With K. Landsteiner and W. Pauli it is well to regard antigens and antibodies as amphoteric electrolytes. The difference between bacteria and bacteria bearing agglutinin is clearly demonstrated (H. Bechhold*i, M. Neisser and U. Friede- mann*). Though the former (typhoid, dysentery) migrate to the anode, the latter lose their charge on account of the agglutinin and precipitate between the electrodes. More characteristic than electrical migration is the behavior of toxins, antigens and immune bodies to acids and alkahs.^ From our standpoint, only very weak dilutions of H and OH are considered from such as cause no irreversible destruction in the substances affected. Acidity diminishes the toxicity of some toxins, but it is restored by neutraHzing them. Kirschbaum isolated a nontoxic dysentery toxin by ultrafiltration and precipitation with acids; it dissolved in alkalis and forms a toxic salt-like combination. An observation made by J. Morgenroth *2 points to the occurrence of a salt-like combination of cobra hemolysin and also of cobra neuro- toxin; these neutral toxins, although colloids, diffuse through an animal membrane into a solution containing hydrochloric acid. Cobra venom may be recovered from crystalloid cobra venom salt by neutralizing it, yet it gradually goes into the colloidal state as I 1 P. Schmidt assumes an additional phase; modification of the bacteria by agglutinin, adsorption of globulin by modified bacteria, flocculation by elec- trolytes. 2 Only great dilutions of H and OH are here considered, such as do not cause an irreversible change in the material. 206 COLLOIDS IN BIOLOGY AND MEDICINE have inferred from the following experiments of J. Morgenroth and D. Pane.*5 I wish to call attention to one other property which is strongly suggestive of colloids. J. Morgbnroth and D. Pane* heated cobra venom in n/20 HCl solution and determined its hemolytic action immediately after cooling and neutraUzation. The hemolytic activity induced by lecithin was greatly diminished but gradually (after hours or days) resumed its original strength. It seems reason- able to regard the gradual restoration of toxicity as phenomena of "maturation" since the particles of the molecular dispersed cobra hemolysin, gradually unite to larger agglomerations and thus ac- quire greater adsorptive capacity. Colloid chemistry offers numerous similar examples; to mention only the aging of dye solutions (hemotoxylin) which must occur pre- vious to its utilization in histological stains. The same interpreta- tion applies to the anologous observation upon the neurotoxin of cobra venom. As a rule the union of antigen and immune substances is inhibited both by H and by OH ions. Just as H and OH ions may break down the union of toxin and antitoxin so may they dissolve the bonds holding agglutinin to its substrate (Hahn and R. Trommsdorf), or either abrin or amboceptor to blood corpuscles. The influence of reaction on the action of hemolytic sera is told in the researches of S. Abramow, Hecker, L. von Liebermann, P. RoNDONi, H. Sachs and Altmann, L. Michaelis and Skwirsky (see P. RoNDONi* for bibliography). Under certain conditions hemolysis is hastened by slight acidity and retarded by larger quantities of acids or by alkahs. The inhibiting action of alkah and to a less definite degree, of acid, is evident in antigen-antibody combinations, as is revealed by com- plement deviation (see p. 207). Its significance is also evident in the Wassermann reaction. Addition of 1/1000 to 1/3200 normal NaOH may inhibit the reac- tion in a strongly positive serum; similarly, a negatively reacting (luetic serum) may give a strongly positive reaction after the addition of 1/1000 to 1/2000 HCl (H. Sachs and Altmann). The following findings favor the view that the physical fixation of amboceptor by blood corpuscles is influenced by the reaction. By means of alkali, the blood corpuscles may be prevented from com- bining with the amboceptor; on the other hand amboceptor-laden blood corpuscles may be deprived of amboceptor by alkahs, and the amboceptor may be recovered in an active condition. The facts for acids are not so obvious. IMMUNITY REACTIONS 207 Complement Fixation and the Wassermann Reaction. A mixture of antigen and immune substance {e.g., goat serum + goat-rabbit serum) has the property of binding complement. This is recognized by the following: if complement is added to a suspension of red blood corpuscles + amboceptor, hemolysis occurs. If the complement has previously been mixed with antigen + immune substance and we then add the entire mixture to the blood corpuscles + amboceptor, no hemolysis occurs. Complement Complement + + Amboceptor Amboceptor Amboceptor + Antigen + + + + Erythrocytes Erythrocytes Erythrocytes Immune substance Hemolysis No Hemolysis No Hemolysis This phenomenon, which is called com'plement fixation or com- plement deviation, was discovered by 0. Gengou and Bordet. It has acquired great practical significance by its utilization for the recognition of antigen by M. Neisser and H. Sachs (one billionth c.c. of human blood may be recognized by complement deviation) and also indirectly, through a reaction analogous to the Wassermann reaction for the recognition of luetic infection. A mixture of antigen and immune serum may give a precipitate, though only when mixed in definite proportions, otherwise this does not occur. Complement fixation occurs regardless of the occurrence of a precipitation. Since complement is easily adsorbed by many colloids and suspensions, it was natural to suppose that the precipi- tate of antigen and immune substances was the fixing agent. This view is held especially by U. Friedemann who attributes to eu- globulin the complement binding power of the immune serum. The investigations of Dean are of great interest to students of colloid chemistry. According to these investigations, much complement is bound when there is a slowly developing turbidity and but Kttle when turbidity develops rapidly. From this aspect a definite develop- ment of surface tension favors binding of the complement. Though it is probable that binding of complement depends on a physical ad- sorption of the visible or invisible precipitate, its mechanism requires further elucidation. The objection that the fixation may occur even without the appearance of a precipitate cannot definitely be proven. We know that albumin particles may aggregate into larger particles without a precipitation, provided the excess of one of the precipitate- forming colloids acts as a protective colloid. On the other hand, it 208 COLLOIDS IN BIOLOGY AND MEDICINE has not yet really been demonstrated that a physical fixation and not an irreversible chemical change occurs in complement fixation. Therefore, in what category complement fixation by means of anti- gen plus immune substance is to be placed, is still an open question. Wassermann Reactjon. Complement fixation is much more obvious in the Wassermann reaction. When A. Wassermann began his studies, he started with the assumption that extract of spirochsete (as antigen) -|- luetic serum (as immune substance) must fix complement. It was very soon evident, that spirochsete extract could be replaced by numerous lipoid suspensions: — by lecithin (0. Purges and Maier), by sodium glycocholate (Levaditi and Yamanouchi), vaseline (Fleischmann), by sodium oleate (H. Sachs and Altmann), sodium palmitate and stearate (P. Hessberg), potato extract and emulsion of shellac (Munk). As Elias, 0. Neubauer, 0. Purges and Salomon showed, these hydrophile lipoids give precipitates with the globulin of luetic sera which fix complement.^ The parallelism between precipitation reaction and complement fixation is very suggestive of a colloid phenomenon. It would be a convincing proof that complement fixation in the Wassermann reaction was a surface phenomenon, if it could be demonstrated that the reaction did not occur in the absence of a precipitation. This proof I gather from an observation of H. Sachs and P. Rondoni. They found that complement fixation by an alcoholic extract of syph- ihtic livers^ depended upon the manner of dilution with saline solu- tion; diluting drop by drop, they obtained a fluid which bound complement strongly. If the extract is added rapidly to the saline solution, there occurs a weak complement fixation or none at all. By slowly adding drop by drop we obtain a turbid fluid; by rapid mixing a clear fluid. We have here two fluids, which in their ability to fix complement can be distinguished only hy the surfaces of the suspended lipoids. The observation of F. Munk also confirms this view that only alcoholic or acetone and not ethereal extracts or solutions of the above-mentioned lipoids are suitable for binding of complement. 1 The observations of these authors are very interesting. They observed that even normal sera give precipitates with the hpoids, but that the range of preci- pitation is much broader with luetic sera and that the complement fixation pre- supposes an optimum mass relationship of lipoid and serum. 2 Before it was discovered that the above-mentioned lipoids were suitable for the Wassermann reaction, extracts of syphilitic livers were employed. IMMUNITY REACTIONS 209 P. Hessberg observed^ that freshly prepared solution of sodium palmitate bound complement, but it lost this property by repeated heating. This change by means of repeated heating is a character- istic colloid property, which evidently is associated with a fragmen- tation of the particles; the more frequently gelatin, agar-agar, etc., are heated, the more difficult it is to solidify them. Unfortunately P. Hessbeeg did not determine whether the repeatedly heated solu- tion of sodium palmitate recovered, on long standing, its ability to fix complement. ANAPHYLAXIS, DEFENSIVE FERMENTS, AND MEIO- STAGMIN REACTION. Anaphylaxis. If an animal {e.g., a guinea pig) is injected with antigen (e.g., horse serum) there are no sequelae. If the injection is repeated after an interval of about 10 to 14 days there occur serious symptoms of poison- ing (convulsions, rise of temperature and respiratory distress) which frequently terminate fatally in a few minutes (anaphylactic shock). This condition induced by the first injection of serum or bacteria is called anaphylaxis (induced defenselessness — the reverse of im- munity). Human "serum sickness" is also a phenomenon of ana- phylaxis which appears after the repeated injection of curative sera and manifests itself in erythemata, swelling of the lymph nodes and joints and moderate rises of temperature. Anaphylaxis is strongly specific, which means that it occurs only upon repeated injections of the same protein or the same strain of bacteria. The specificity is so absolute and the quantities required so minute, that like precipita- tin reactions, anaphylactic phenomena may be employed in dis- tinguishing traces of human from animal blood or in detecting adulterations. Friedberger and his coworkers were successful in preparing the anaphylactic poison (anaphylatoxin) outside the body, in vitro. If an animal (e.g., a guinea pig) is injected with antigen (e.g., horse serum) antibody appears in the blood after a time. Friedberger proceeded from this fact; he mixed antigen and antibody in a test tube and obtained a precipitate from the mixture. By digesting the precipitate with guinea-pig serum (which always contains com- plement) he obtained what he designated as anaphylotoxin. Since peptones produce phenomena resembhng anaphylactic shock, it was thought that peptone-like products were split off, perhaps by fermentation in the interaction between antibodies and antigen. 210 COLLOIDS IN BIOLOGY AND MEDICINE This view, still held by Friedberger, regards the appearance of the anaphylactic poison as the result of a fermentation of antigen by the degradation action of complement resulting from antibody fixation. It was a surprise to find that the anaphylatoxin might be prepared even without antibodies as when guinea-pig serum was digested with bacteria; and even by digestion with colloidal carbohydrates (agar, starches, starch paste, pectin and inulin).* [The work of Vaughan and NovY on the protein poisoning should be mentioned as well as that of R. Weil on the mechanism of anaphylaxis. Tr.] H. Sachs and E. Nathan demonstrated that the physical state of the poison-producing substances was a determining factor. They employed inulin as absorbent. This carbohydrate is practically insoluble in cold water though it forms a clear solution without pasting in warm water. Upon mixing guinea-pig serum with a 5 per cent suspension of inulin, a serum was obtained which produced the severest anaphylactic shock when injected into guinea pigs, though no toxic substance resulted from the mixture with inulin solution. Pastes have a very extensive surface development, consequently, anaphylatoxin is formed best with starch paste. H. Sachs and E. Nathan find in these experiments decisive confirmation of the physical theory of anaphylaxis first proposed by H. Sachs and Ritz. According to them, antigen is not the mother substance of anaphyla- toxin, which is not newly formed, but which exists preformed in normal serum. This poison becomes active by the adsorption (separation) of a substance as yet undetermined. In the case of artificial anaphylatoxins, bacteria and carbohydrates serve; in true anaphylaxis, the products of antigen and immune body act as ad- sorbents« The specificity of anaphylaxis is thus explained since only the specific antibody formed causes a orecipitate with the antigen. Protective Ferment. The remarkable relation between immunity reactions and pro- tective ferments should be mentioned here. E. Abderhalden con- siders under this term the ferments which destroy and render innocuous species-foreign proteins, entering the organism parent- ally. Since the connection between protective ferments and colloid research is still unestablished we shall merely refer to the work of E. Abderhalden.^ 1 It is not yet definitely established that anaphylatoxin may be made by shaking inorganic suspension with normal guinea-pig serum. 2 [D. Van Slyke has thoroughly discredited the Abderhalden reaction with his nitrous acid method for the quantitative determination of amino acids. Harvey Lectures, 1915-16, p. 170, Lippincott, N. Y. Tr.J IMMUNITY REACTIONS 211 The Meiostagmin Reaction. M. AscoLi and G. Izar found that substances which lower surface tension of the solution are formed in the reaction between antigens and immune bodies. He determined this with Traube 's stalag- mometer by counting the drops which formed from a measured quantity of fluid.^ When, for instance, he mixed an extract of typhus bacilli with normal serum and with the serum of typhoid patients, 10 c.c. gave 58 drops in the former instance and 61 drops in the latter. M. Ascoli considers it to be a general reaction and has employed it in the serum diagnosis of various conditions (syphiHs, tuberculosis, anchylostomiasis and echinococcus infection). It has not been generally introduced as a means chnical diagnosis of infec- tious disease since the technic is so precise that the differences are within the limit of error, but it has been more frequently employed in the detection of malignant growths. Two dilutions of the serum are prepared, one with water and the other with an equal quantity of tumor extract (recently Ascoli employed ricinoleic acid or linoleic acid, etc.). If the number of drops are larger in the latter mixture than in the former, there is a presumption that the serum is from a cancer patient.^ 1 Meiostagmin. reaction = reaction of smaller sized drops. 2 [E. P. Bernstein and Irving E. Simons, after critically reviewing the litera- ture and their personal experience, have discarded the Meiostagmin reaction as useless cUnically. Amer. Jour. Med. Sei., vol. 142, p. 862, et seq. Tr.] An asterisk (*) after an author's name refers to a reference in the index of names. PART III. THE ORGANISM AS A COLLOID SYSTEM. The Significance of the Colloidal Condition for the Organism. Recently, I read in a French magazine, a fantastic description of a visit to the Martians. They were pictured as men mth iron faces; with a great bill replacing the nose; three glass eyes and joints and limbs of iron. VThj did not the artist construct his people of a material actually found on that planet? If we assume that life exists on other planets and disregard the theory of pan- spermogenesis ^ accepted as probable by Sv. Aerhexius, a priori it is probable that life is associated with substances similar to those with which it is associated on our earth. One thing I can assert, that whatever the material comyosition of such liimig beings may he, it must he colloidal in nature. Those iron Martians could no more exist than could crystallized life. What condition of matter, other than the colloidal, could adopt such changeable, such plastic shapes, and yet, when necessary, be in a position to maintain them. An exchange of substance may occur in jellies as well as in a fluid; in the latter the least touch, an miintentioned movement, disturbs the result of diffusion and brings about the death of the system; the changes in a jell 3" are fixed as m a solid mass. Colloids may form permeable walls or membranes, whose permeability is regulated by the substances which pass through them; thus, for instance, the sulphates which are less important to the organism close the passages on themselves; the chlorids facilitate their otsti entrance. Foods enter our digestive tract in colloidal condition, as albumin and starches. Made fluid and easily diffusible bj^ the enzjTnes, they penetrate the organism in order to be fixed and again transformed into colloids. In that condition only are they retained by the or- ganism and prevented from flowdng away. Colloids, because of their ^ According to this theory, it is conceivable that germs of life travel from one planet to another and that they develop there under favorable circumstances, so that, to a certain extent, one star infects another one with life. 213 214 COLLOIDS IN BIOLOGY AND MEDICINE surface development, unite the advantages of the soHd condition with that of the fluid; observe for a moment a mountain cUmber, a fever patient, or a tree in the springtime, which is decked with leaves in four or five days. What enormous amounts of chemical energy are expended by the mountain climber in a few hours, what large quantities of protein are in a short time consumed by the fever patient, what large quantities of material are carried to and from the periph- ery of the tree. With the least loss of time the reserves must be mobilized and carried to the seat of war, the places where they are consumed. Such a rapid mobilization is unthinkable in the case of a solid crystalloid, with its small surface; a chemical process in a swollen colloid may only occupy minutes, whereas the same reaction in a shrunken colloid requires days. How wonderful by means of adsorption is the action of surface development as a regulating mechanism. Luxus consumption of foods, salts, oxygen, etc., are excreted as quickly as possible by the colloid components of the body, but when the supply ceases, the amount given off becomes less and when there is a deficiency the organism tenaciously retains the last traces for its time of need. Quantitatively, the substance most important for the organism is water; colloid and water are one in the organism; an organism without water is lifeless. We can imagine such an intimate and varying relation to water only in a colloid system; the process of swelling, the adsorption of water, and shrinking to complete dryness exhibit no leaps or sudden changes in condition. If we compare crystalloids with colloids, we shall see that something entirely new with very changed properties, a solid crystalloid precipitate, appears from a solution upon losing water. Such a system would be unable to maintain correctly the constantly oscillating water balance and the normal condition of swelling in the organism, or to act as an accumulator of large quantities of water, like muscle, and release it for use when necessary. Such a system could not, like a pen, smooth out the irregular chemical impulses, which the organism experiences as the result of physiological and pathological life processes, and which, after absorption, constantly restores its state of swelling to normal by means of secretion (kidney, skin, etc.). We thus see that the processes which cause us to marvel at the wonderful adaptability of Nature rest upon the simple laws applicable to colloids. Thus it is that I am unable to imagine that the com- plicated and adaptive phenomena of Life could possibly be associated with any other than a colloidal system. CHAPTER XIV. METABOLISM AND THE DISTRIBUTION OF MATERIAL. The Distribution of Water in the Normal Organism. The earliest stages in the development of life are accompanied by powerful processes of swelling which soon reach a maximmn, and then pass over into a shrinking, which becomes progressively greater, until death occurs. In the cases of plants, the struggle for water be- tween seed and soil starts with germination (A. Muntz*). Growing and full-grown plants show a certain turgor, i.e., a fulness or tension like a distended rubber balloon, while a dying plant is withered and poor in water. A three months' human fetus contains 94 per cent water; at birth the water content is from 69 to 66 per cent; in adult life 58 per cent.^ I am not acquainted with any figures of the water content of the aged, but it is generally and with justice assumed, that in old age, the water content decreases; turgescence in general, and of the skin in particular, is obviously lost. The organism, taken as a whole, dur- ing its life evidently passes through the curves of swelling and shrink- ing of an inelastic gel. In individuals of the same species, the water content is probably fairly constant for the same period of life. The water content of the individual portions of plants and of simi- lar organs of different plants varies remarkably. Though jelly-like protoplasm contains from 60 to 90 per cent of water, the dry wall of ligneous cells takes up from 48 to 51 per cent, while the jelly-like membranes of nostocacese and palmellacese absorb as much as 200 per cent of water, according to NÄgeli; on the other hand, cork membranes have hardly any swelling capacity at all. The resistance offered to loss of water is exceptionally variable. It may be said, with certain exceptions, that plants are much more resistant than animals. Especially the lower forms of life, particu- larly the spores of bacteria, yeasts, algse, mosses and seeds may bear almost complete dehydration without dying. Loss of water is often of great biological significance for plants. It makes spores and seeds less sensitive to changes of temperature; and in the case of some ^ The greater water content of the individual organs of the newborn as con- trasted with those of adults is especially evident from the tables of E. Bischoff.* 215 216 COLLOIDS IN BIOLOGY AND MEDICINE higher plants, the spreading of the seed pods upon drying leads to movements which serve to distribute the seeds. Best known is the "blooming," or the swelling of the "Jericho-rose." Higher animals on the other hand are very sensitive to losses of water: frogs, according to Kunde, may withstand a gradual loss of water up to 30 per cent, but, if they are rapidly dried, they perish when the loss is only 18 per cent. In the latter case, there is evidently no time for an equali- zation in the distribution of the water. Thirsting human beings also show great losses of water, though of course, there are no data as to the lethal point. A. Durig informs me that, after a forced march in hot weather, he lost 5 kg. of water. The investigations of N. ZuNTZ and Schumberg on marching soldiers, as well as those of N. ZuNTZ on mountain climbers, showed that exercising men lost water. Roughly measured, the water ingested after forced marches does not replace the water lost. We may say here, anticipating some- what, what animal experiments of H. Gerhartz* show, that the loss of water affected primarily the musculature and then the fluids cir- culating in the organs. Freezing (gefrieren)^ has an effect on the organism similar to the withdrawal of water. It was formerly believed that the formation of ice burst the cell walls or tore the protoplasm, and the damage from freezing was ascribed to these gross mechanical influences. It was shown by the investigations of A. E. Nägeli, W. Sachs, H. Molisch, and Müller-Thurgau that these views were false, that usually there was no formation of ice in the cell, but that the ice crystals grew between the cells in the intercellular spaces. P. Matruchot and MoLLiARD* studied plant cells and found that the phenomena observed in drying or plasmolysis resembled those induced by freez- ing (erfrieren). H. W. Fischer,*^ as the result of exhaustive studies, reached the conclusion that the damage done to animals and plants by freezing them (gefrieren) was analogous to the partially irre- versible changes produced in gels by glaciation. He believes that the adsorption of electrolytes in particular is thus affected unfavor- ably. If a solution of potato starch is frozen and then thawed out, the electrolytes may entirely dissolve again but the starch has become insoluble. Frozen leaves present an analogous condition in that the chlorophyl is no longer retained. In this valuable work he 1 Freezing (erfrieren) and glaciation (gefrieren) must not be confused. A plant or an animal freezes if life processes cease by reason of the low temperature. This temperature depends upon the nature of the organisms involved, and in the case of warm-blooded individuals is usually far above zero; in the case of other organisms (seeds and spores), however, it may be far below zero (—200). Glaciation always means ice formation. METABOLISM AND THE DISTRIBUTION OF MATERIAL 217 shows that the condition and age of protoplasm when frozen, i.e., at the lethal point, is similar to that observed by van Bemmelen in the drying of colloids; when the latter are frozen, they become optically inhomogeneous, and their staining capacity changes. For normal functioning there is an indispensable normal water content for every individual or- ganism and organ. The total water content of an organism gives us an idea of its water requirements; the water content of the individual organs informs us about the distribution of water in the organism. A clearer idea is obtained by observing the distribution of water in animals especially in mammals. In Table III (see p. 219), I have compiled the water content of the different adult organs from available data. Table II (p. 218) shows the distribution of the total water content (100 per cent) in the various organs. I have placed the proportionate weights of the organs to the total weight alongside for com- parison. The data in Tables I and II show far-reaching differ- ences (see especially Skin in Table II), for which age and nu- tritive condition are responsible. In healthy animals and men, there is a definite swelling ratio for the individual organs, a dynamic equilibrium. The maximal varia- tion of normal swelling, more or less, is called the swelling range.^ It 1 Definitions: Swelling capacity is the maximal capacity for absorbing water, expressed in W (weight of water) D (dry weight) Swelling is the water content of a gel or an organ expressed in W D ' Swelling range is the greatest variation in the capacity of an organ to absorb water under different conditions. • W (maximal weight of water) — W (minimal weight of water) D ■ A normal swelling range and an abnormal swelling range exist. I a 6 c Fig. 37. Spirogyra: a, before the ex- periment; b, frozen; c, after thawing. (From H. Molisch.) 218 COLLOIDS IN BIOLOGY AND MEDICINE TABLE II. Percentage Distribution of Water in the Individual Organs. (Entire water in the organism = 100 per cent.) (After A. Albu and C. Neuberg,* Bischoff,* Engels * and A. W. Volkmann.*) Human. Dog. Distribution of total water in each organ. Weight of organs in per cent of body weight. Distribu- tion of water in each organ. Weight of otgans in per cent of body weight. Blood 4.7—9 2.3 6.6—11.0 3.2 2.8 2.4 0.4 0.6 9—12.5 47.74—50.8 2.7 4.9 Newborn 13.5 Man 18.2 Woman 28.2 Newborn 11.3 Man 6.9 Woman 5 . 7 Man 4.1 Woman 5 . 4 8.27 11.58 9.68 3.86 2.83 1.01 9.08 47.74 1.59 7 Fat Skin 16.11 Viscera: Intestines 8.18 Liver 3.60 2.36 Spleen Kidneys 0.85 Bony skeleton Muscles Newborn 15.7-17.7 Adult 15.9 Newborn 22.9-23.5 Man 41.8 Woman 35.8 Newborn 12.2-15.8 Man 2.6 Woman 2.7 17.39 42.84 Nerve substance: Brain Spinal cord . . . . 1.37 Remainder 11.0 is very small for the skeleton, the blood, and the intestines, has a middle value for the viscera, and becomes higher for skin, muscles, and kidneys. This was deduced especially from the experiments of Engels.* He kept dogs without food four days and determined the water content of various organs (normal animals). Another series of dogs, after the same preUminary treatment, received an infusion into their jugular veins of 1160 gm. physiological salt solution (on the average). Three hours after the termination of the infusion, the animals were killed and the water content of the organs determined (water treated animals). In the following table, the percentage of the water infused, that is, recovered from the individual organs, is shown in column A, and in column B is the percentage increase in weight of the several organs in terms of their own weight : METABOLISM AND THE DISTRIBUTION OF MATERIAL 219 (Basis of the swelling range.) A. ß. Muscles 67.89 17.75 2.96 2.25 1.97 1.55 1.41 1.13 0.28 2.82 100.01 17.1 11.9 8.9 3.0 9.0 2.4 17.9 8.9 10.0 Skin Liver Intestine Lungs Blood Kidneys Brain Uterus Lost in bleeding TABLE III. Water Content of Various Human Organs in Per Cent.^ (After Albu-Neuberg,* Bischoff,* Halliburton,* Pfibram,*i Rumpf.*) Adult (normal). Child (normal). N — Newborn. 2 M — 2 months old. Adult (pathological). Blood Fat 77 9— 83 29.9 73.3—77 79.2—80.2 68.3—79.8 78—79 75.8—86.1 77—83.7 22—34 73—75.7 75—82 N85 N83.1 2 M 75.5 N 83.3—93.1 2M80 N 80.5 2M73 N82.6 2 M 79.4 N78.4 2 M 77.7 N 85.7 2M81 N32.3 2 M 62.3 N81.8 2M71.7 N89.3 2M89 90 and more (Anemia) 75.5—83.9 (Nephritis) 73.2—66.5 (Diabetes) Viscera: Intestines Heart 79.2—80.4 (Nephritis) 68.5—87.3 (Nephritis) Liver Lung Spleen 90.6 (Nephritis) 84.8—88.2 (Nephritis) • Kidneys Bony skeleton Muscles Nerve substance: Brain 80.9—83 (Nephritis) 1 I have omitted the figures for skin; they vary for different authors between 31.9 and 73.9, be- cause some have given the water content of skin deprived of fat and others that with the fat attached. Since, in the dog, muscles constitute 42,82 per cent and skin 16.11 per cent of the total body weight, these organs under normal con- ditions actually store up, respectively, 47.74 per cent and 11.58 per cent of all the water in the body. As a result of their great swelling range, tlie two chief water excreting organs, the skin and kidneys, the 220 COLLOIDS IN BIOLOGY AND MEDICINE muscles, most of all, are able to accumulate large quantities of water. In Engel's experiment they took up 2/3 of the water supplied. If water is suppHed to the organism, the blood, on account of its low swelling range, gives off the excess chiefly to the muscles and skin; it parts with some to the glands, chiefly the kidneys. On this account, saline infusions after a severe loss of blood have usually only a temporary effect.^ Muscles and skin behave like a reservoir, the blood, like a rigid pipe system, from which, if the pressure is sufficient, excess of water constantly flows through a small vent. In carrying out this wise arrangement, the organism utilizes the various swelling ranges of the organ colloids. Though I have used the above picture of a rigid pipe system for the blood, it is not strictly accurate, for the blood has a small swelling range of its own, as we see from Engel's table. This may be attributed to the fibrinogen, as we learn from the following facts. The required data I have taken from E. Abderhalden (pp. 592-593). *i The water content of various animals is: Per mil. ' In the entire blood 749-824 Serum 902-926 Blood corpuscles 604-633 From this we see that when the water content of serum increases 2.6 per cent (from the minimum), it reaches its maximum, and the blood corpuscles reach their maximum with an increase of 5 per cent. The entire blood on the contrary has a swelling range of 10 per cent. There must therefore be something in the blood that swells especially well and the only possible substance is the fibrinogen. Let us compare the maximal and minimal content of water dis- tributed between serum, blood corpuscles, and the whole blood in the identical animals. Max. = maximum water content among the various species of animals. Min. = minimum water content among the various species of animals. ^ Attempts to hold the fluid in the vessels by the addition of colloids have been unsatisfactory. A more favorable result is obtained when 14 grams of salt and 10 grains of crystalline sodium carbonate are administered either intrave- nously or by rectum (J. J. Hogan and M. H. Fischer). It was accomplished through reducing the swelling of the other tissues by hypertonic saline and neu- tralization of acid by the alkali. Cholera collapse, which results from the water deprivation by reason of diarrhoea, may be successfully combated by hypertonic saline infusions (Roger)-. [W. M. Bayliss and M. H. Fischer have recommended the use of gum arable solution, and it is being successfully employed at the front in the present war, see p. 137. Tr j METABOLISM AND THE DISTRIBUTION OF MATERIAL 221 Serum Corpuscles. Entire blood. Cat Horse I Sheep I 926Voo (max.) 902Voo(min.) 917Voo 925Voo 624Voo 613Voo 604Voo(min.) 633Voo (max.) 795Voo 749Voo(min.) 821Voo Rabbit 817Voo The serum of the cat had a higher water content than that of the other animals experimented upon; the corpuscles also had a high water content, even though not the highest; the entire blood, how- ever, is only a little above the average. For Horse I, serum and entire blood have minimum values and the corpuscles a low but by no means the lowest value. In the case of Sheep I, the entire blood reaches a maximum, whereas the blood corpuscles show a minimum, and the serum possesses a water content a little above the average. With the rabbit, there is a high water content in all portions. From this we learn that a substance possessing a great range of swelling exists in the whole blood, namely, the fibrinogen. We recognize further, that the elements of the blood possess a cer- tain elasticity, which smoothes out the fluctuations in the water content, and which shows itself by a " Zog " of the water plenitude or poverty in the various elements of the blood, depending on whether there is a supply or a withdrawal of water, a swelling or a shrinking. The following swelling ranges are found for the various elements of the blood: fibrin > whole blood > corpuscles > serum. It would be desirable to have investigations of the water content of the differ- ent elements of the blood in the same animal before and after water has been given. What is it then, that determines the water content or swelling of an organ? Undoubtedly each organ colloid has a definite swelling ca- pacity and a definite swelling range. A priori, we may assume that the colloids of muscles swell more than those of the epidermis. Without doubt, the structure of the given colloid is also a factor. W. Pfeffer*^ {loc. cit. I, p. 61) justly emphasizes the distinction between water of swelling, consequent upon the hydrophile state of the swelling substances and the water of imbibition, which is drawn up into the capillary interstices as into a sponge.^ E. Pribram*^ beUeves that the sweUing of protoplasm in its true sense, i.e , of the assimilated (species-native) colloids of the cell is con- 1 W. Pfeffer speaks, it is true, of "molecular" water of imbibition (or ad- herent water) and of "capillary" water of imbibition, yet he intends the same distinction that I have indicated above. 222 COLLOIDS IN BIOLOGY AND MEDICINE stant. Only the swelling of the nonassimilated reserve substance is variable. This view, it seems to me, receives its chief support from the findings of H. W. Fischer and P. Jensen* on muscle, which is treated more thoroughly on page 291. According to their findings, water occurs in muscle in two phases. One has a constant value and is closely associated with the viability of muscle (conditioned by the integral muscle protoplasm). The other phase varies in the ex- ercising muscle and is only loosely bound (water of swelling of the reserve material). Besides these factors of swelling which are inherent in the organ colloids under consideration, there are others which are to a certain extent impressed from without. The natural salt content as well as the products of metabolism, especially the acids, are determinative of the swelling of a tissue. Acid formation in an organ {e.g., CO2 or lactic acid in active muscle; CO2 in blood corpuscles) increases its swell- ing capacity. If we observe that the potassium salts predominate in one organ, and in others, the sodium salts, or even that there is an accumulation of salts in a certain portion of a single cell, we may conclude from that alone, that the water content also depends upon such concentration. Potassium salts increase swelling; Ca salts deplete (E. Widmark*), and according to R-. Chiari and Januschke inhibit exudation. When the loss of water is very great (cholera, infant diarrhoeas), there is an increase of potassium salts and phosphates in the urine. From this it may be assumed that Na salts replace the K salts of the muscles, and at the same time water is given off. According to E. Pribram, *2 this occurs in order to protect more vital organs, espe- cially the brain, from loss of water. Little is known concerning swelling from a biological point of view. On this account, an observation of H. Paul* seems especially noteworthy. He pointed out in the case of peat mosses that the high moor sphagnum is able to absorb much more water than the low moor sphagnum. For example, sphagnum molluscum absorbs 27 times, and sphagnum platyphyllum absorbs 16 times its dry weight of water. In the same paper we find that the high moor sphagnum contains much more acid than the low moor sphagnum, and that the former are much more sensitive than the latter to the action of alkalis, lime and salt. From this it seems to me we may certainly infer that swelhng of high moor sphagnum is greater than that of low moor sphagnum because of its greater acidity, and that the damage it suffers from salts, etc., may be attributed to the alteration in its normal condition of swelling. We shall see that abnormal accumulation of acid in the tissues METABOLISM AND THE DISTRIBUTION OF MATERIAL 223 results in their swelling or edema. We recognize from this, that the dynamic balance of the swelling is dependent upon the normal course of the processes of assimilation and dissimilation. Con- versely, pathological processes are always followed by an abnormal condition of swelling. Pathology of Water Distribution. In pathological conditions, the water content may have values very different from normal. The water in the blood rises to 90 per cent and more in severe anemias, and may fall to from 73.2 to 66.5 per cent in diabetes (see p. 219). Under other pathological condi- tions, organs may show abnormal swelling (see last column in Table III, p. 219). With the active metabolism and formation of crystalloids, which occur in fever, there are alterations in swelling (thirst, dryness of the skin), the exact nature of which we do not as yet understand. As a rule, there has been less attention paid to the study of the conditions in which the swelling of an organ is below normal. The injection of protoplasmic poisons (some heavy metal salts, strong acids) causes a coagulation of the organ albumen, which reduces to a greater or less extent its swelling capacity. I am still occupied with more exhaustive studies of these questions which are also touched upon in the chapter on Necrosis. It is too early to report the results. Edema. By edema we understand an abnormal collection of fluid in tissue or tissue spaces; if the fluid collects abnormally in a body cavity, we call it an exudate or hydrops. The view most generally accepted up to a few years ago was that edema occurred whenever the venous blood pressure, or more cor- rectly, the difference between arterial and venous pressure was gen- erally or locally raised, and the resistance of the vessel walls was diminished (Julius Cohnheim, 1877). It is known that in heart disease and in nephritis, when the circulation is disturbed, that large portions of the body, especially the lower extremities, swell and become edematous. The local inflammatory edema accompanying inflammation;; insect;bites or the injection of an irritating fluid (e.g., diphtheria toxin) must also be considered. The above explanation- has some fascination, and it cannot be denied that it will continue to be invoked in explanation of certain points, especially since the observation of increase in the permeability of vessel walls cannot be avoided, when we see that even corpuscles 224 COLLOIDS IN BIOLOGY AND MEDICINE I'iG. 38. The ligated leg (right) of the frog is edematous. (From M. H. Fischer.) Reproduced from Fischer's " Oedema and Nephritis," second edition, by permission of Messrs. John Wiley & Sons, Inc. METABOLISM AND THE DISTRIBUTION OF MATERIAL 225 pass through the walls. In general, however, it may be asserted that the above explanation has brought no advance to our under- standing of edema. Histologists and physiologists have contributed an interminable amount of work without making any progress. A fundamental departure in the understanding of edema and its associated questions was made in 1907 by the investigations of an American, Martin H. Fischer^, who sought the cause of edema, not in the vessels, but in the tissues themselves; he attributes to the edematous tissues an increased swelling capacity. M. H. Fischer's views were immediately contradicted and led to more experimental studies and scientific discussions than most biological theories. However, though we must now admit that M. H. Fischer's views are too far reaching, it cannot be denied that he set up a very productive working hypothesis. We shall first state his theory and then discuss his opponents' views. The most important experiment of M. H. Fischer is the follow- ing : he ligatures the hind limb of a frog so that its circulation is cut Fig. 39. Rabbit's kidneys; left normal, right experimentally edematous. off (Fig. 38) and places it in water so that the limbs are covered. The hgatured limb swells up and at the end of 2 or 3 days may be 2 or 3 times its original weight. If the frog is kept in a dry vessel, the ligatured limb dries up completely, and if it is cut off and placed in water, it swells up. Under these conditions the blood pressure or the increased permeability of the vessel walls cannot play any part in the development of the edema, but it is only the tissues, which swell more strongly under the circumstances mentioned. In the same manner, M. H. Fischer was able to demonstrate the occurrence of edema in rabbits' kidneys (Fig. 39), and in the livers and lungs of ^ He was led to this by experiments of Jacques Loeb, which showed that frogs muscles swelled more in acid and alkaline fluids than in neutral ones. 226 COLLOIDS IN BIOLOGY AND MEDICINE f sheep. These became edematous, not only if the veins were hgatured, but also if the artery was tied. Under the circumstances, an in- creased blood pressure as the result of congestion need not be con- sidered. Indeed, all dead bodies or portions of them which certainly are without blood pressure swell up when immersed in water. It was thus demonstrated that the development of edema de- pended upon the increased capacity of the tissues to swell, and the question then presented itself, What change in the tissues permits the development of edema? An explanation was provided by the studies of F, Hofmeister, his pupils and successors, upon the swelling capacity of gelatin and re- lated substances (see pp. 67-68). We learned in that chapter that acids and alkahes increase swelling capacity, that other electrolytes in the order there mentioned, either favor or diminish swelling, while nonelectrolytes, as far as appropriate studies as yet show, have only slight influence. Martin H. Fischer* (in collaboration with Gertrude Moore) was able to demonstrate that the same laws governed the swelling capacity of fibrin, the swelling of frogs' muscles and the extirpated eyes of oxen and sheep. This explana- tion presupposes no membrane or osmotic pressure; it permits an unstrained interpretation of processes which otherwise are explained with great difficulty in the case of animal cells unprovided with membranes. The further question now presents itself. What electrolytes are re- sponsible for the altered swelling capacity of the tissues in edema? We can no longer offer a single explanation, and we must study indi- vidual cases. Edema fluid, the CO2 of which has been removed, is acid to Phenolphthalein; F. Hoppe-Seyler found in it valeric, succinic, butyric and lactic acids. Strassburg and R. Ewald found that the CO2 content of edema fluid was greatly in excess of that of venous blood. ^ Of especial value is the discovery by F. Araki and H. ZiLLESSEN that any lack of oxygen is followed by an excessive production of acid, though this fact may not be demonstrable by indicators. ' One answer is thus given to the question asked above. Increased acid production, one of the results of deficient oxygen, may cause the development of edema. Such a condition occurs in circulatory dis- turbances and in cardiac insufficiency, where edema is especially fre- 1 I wish to point out that a certain contradiction is contained in the simul- taneous presence of organic acids and increased CO2. Possibly this may be attributed to the fact that various edema fluids have been examined and un- justified generaUzations deduced. (The author.) METABOLISM AND THE DISTRIBUTION OF MATERIAL 227 quent; it occurs also in severe anemias and in certain cachexias, starvation and scurvy. In nephritics, some substances recently dis- covered in the blood, may perhaps contribute to the inhibition of oxidation. Cadaveric edema is well known, as is the bloated ap- pearance of drowned bodies. In the case of living animals, only the injection of excessive quantities of water or physiological salt solu- tion are able to bring about edema. According to R. Magnus, this is readily accomphshed by injecting salt solution into the blood vessels of dead animals. A dead frog may double its weight in from 36 to 48 hours if immersed in water. The injection of certain poisons (especially lactic acid) results in an oxygen deficiency and thus brings about an excessive acid pro- duction. M. H. Fischer injected morphin, strychnin, cocain, arsenic and uranium nitrate into the dorsal lymph sac of frogs and thus produced an edema which disappeared if the frog was given an opportunity to excrete the poison. Edema in the case of metal intoxications, especially in the case of arsenic, is well known to clinicians, and the great thirst and the dimi- nution in the excretion of urine which occurs after morphin, ether and chloroform administration, may also be attributed to the deficiency of oxygen in the tissues, with concomitant absorption of water. We have only referred to the development of edema by acids, and I wish to call attention to the fact that intense edema may be pro- duced by alkalies. Subcutaneous injection of n/10 sodium hydrate results in severe edema. If such substances produce edema as favor the swelling of gelatin, fibrin, etc., edema must be counteracted by electrolytes which re- duce swelling. The correctness of this assumption was demonstrated by M. H. Fischer on the amputated leg of a frog. The addition of neutral salts diminished the swelling and acted in the same order that the cations and anions did in diminishing the swelling of fibrin. Nonelectrolytes, on the other hand, had no influence. M. H. Fischer regards glaucoma as a typical example of a local edema. This is a disease of the eyes, of which the most character- istic symptom is very greatly increased tension, which produces hardness of the eyeball. The excruciating pains and loss of vision are mere consequences. The various explanations given in ophthal- mological textbooks are quite unsatisfactory, whereas the experi- ments of M. H. Fischer are quite convincing. M. H. Fischer placed extirpated ox eyes ^ in water, to which so little acid had been added that it was imperceptible to taste. These eyes became stony 1 The investigations of P. Bottazzi* and his pupils on the sweUing and shrinldng of lenses in solutions of acids, bases and salts should be mentioned. 228 COLLOIDS IN BIOLOGY AND MEDICINE hard as in the most aggravated glaucoma. On the other hand, by the injection of a few drops of 1/8 to 1/6 molecular (4,05 to 5.41 per cent) sodium citrate solution under the conjunctiva, it was possible to cause the pressure in human glaucoma and in artificially glaucomatous eyes to become normal or even subnormal in less than five minutes. [The observation of Riesman on the lowered ocular tension of diabetics and the lowering of ocular tension by intravenous injection of glucose by Woodyatt's^ method are important in this connection. Tr.] Cloudiness of the cornea occurs without reference to the absorption of water by the eye. It probably depends on the precipitation of a protein, since all acids, bases, salts and nonelectrolytes which cause a precipitation of proteins cause a corneal turbidity. What is true of the cornea may be supposed to apply also to the other transparent media of the eye, the lens and the vitreous humor. Here, too, therapeutic results have been obtained already, by injecting sodium citrate. Hayward G. TnoivLiS obtained an improvement in vision in cases showing a cloudiness of the cornea, the lens or the vitreous humor. It is natural to assume that the turbidity is due to a reversible pre- cipitation of albumin. Similar results were obtained in an edema of the tissues surrounding the knee. The tiny sweUings which result from insect bites are regarded by M. H. Fischer as local edemas produced by a drop of acid or some substance which interferes with the normal oxidative processes of the tissues. The beneficial action of ammonia, customarily employed on insect bites, favors this view. M. H. Fischer produced "artificial flea bites " on gelatin plates by sticking them with needles dipped in formic acid and then placing them in water; with ammonia he was able to make the swellings recede. One of M. H. Fischer's observations seems of great significance to me in explaining the phenomena of some skin diseases. He ob- served in gelatin plates upon which moulds had been sown, an ele- vation in the centre of each colony (swelling). If we recall that many skin diseases are caused by true fungi related to the moulds, and that in many, the most characteristic symptoms are wheals (swell- ings), papules and vesicles, the analogy cannot be neglected. More- over, we must think of analogous processes if we consider other local edemas occurring after inoculation with infectious microorganisms or the injection of diphtheria antitoxin. In a later work, M. H. Fischer*^ elaborated his fundamental ideas of cellular pathology and studied cloudy swelling. Cloudy swelling is found in the liver, kidneys, spleen, and muscle cells of 1 Studies in Intermediate Carbohydrate Metabolism. R. T. Woodyatt. Harvey Lecture, 1915-1916, p. 328. METABOLISM AND THE DISTRIBUTION OF MATERIAL 229 persons dying as the result of acute infectious diseases. The cells involved are enlarged and more or less cloudy; macroscopically, the organ involved appears at times as if it had been boiled; microscopic- ally, granular deposits are seen. steamy (yellowish') -jviilte white aucleua Fig. 40. Reproduced from Fischer's "Oedema and Nephritis," second edition, by permission of Messrs. John Wiley & Sons, Inc. M. H. Fischer experimentally produced cloudy swelling in the livers and kidneys of rabbits by placing them in distilled water or dilute acids. Addition of various salts delayed or hastened the de- velopment of cloudy swelling. According to M. H. Fischer, whether the swelling occurs pathologically or experimentally, it is explained just as is edema; as the result of the production of acids in the injured tissues, they swell. The cloudiness runs parallel with the precipitation of proteins, especially that of casein by acids. The pro- duction and disappearance of granules in parenchymatous cells in the presence of additional acid occurs exactly hke the precipitation a,nd re-solution of casein upon increasing the concentration of acid. The "cloudy swelling" is, therefore, the result of acid production in the tissues, though the swelling and the clouding are two processes entirely independent of each other. As has been stated already, M. H. Fischer's theory was actively discussed and contradicted. Before approaching this discussion we shall present a brief resume. Fischer maintains that edema results 230 COLLOIDS IN BIOLOGY AND MEDICINE from the swelling of organ colloids, which is induced by acids that are produced by a disturbance in the oxidative processes of an organ. The first objection is directed against the assumption that processes which occur in dead colloids may be transferred to the living or- ganism. It was raised by G. Bentner, Jacques Loeb and A. R. Moore. They admit that Fischer's experiments apply to dead tissue, especially muscle, but that they fail in the living organ where osmotic processes are active and satisfy all the conditions. The objec- tion of R. Holer is especially searching, that if only electrolytes may inhibit acid swelling, a piece of muscle would swell in sugar solution. As a matter of fact the muscle volume is unchanged in an isotonic sugar solution. M. H. Fischer *^ combats this by stating that the existence of osmotic membranes in living cells is quite readily conceivable (see p. 290), and points to his experiments which reproduced the contraction of living muscle by means of catgut, a dead colloid material. Osmotic attraction of water is inconceivable nor does it assist the explanation. The second objection is directed against the very development of acids in tissues and was raised by A. R. Moore, who was unable to detect acids by acid fuchsin or neutral red in muscles made edematous according to Fischer's technic nor in the lymph or kidneys of rabbits injected with acid salts; and who maintained that consequently the acid content of such tissues is not responsible for the edema or albu- minuria. Fischer meets this objection by stating that acidity of tis- sues is not to be detected by color indicators since the acids combine with proteins and that even traces of acids induce swelling and that there is no more delicate indicator of acidity than the swelHng of proteins. [Fischer insists that p^ swelling in protein nowhere par- allels Pb concentration. The degree of swelling follows the order, HCl > lactic > sulphuric acid. Tr.] The third objection is directed against the generaHzation of Fisch- er's hypothesis. Fischer performed his experiments principally on muscles which behaved by swelling in the presence of acids Uke fibrin or other dead colloid but this would not apply to all kinds of tissue. Connective tissue and cartilage apparently behave the same way. L. PiNCUSSOHN, however, found that kidney, spleen, liver and lung usually became less swollen in acid than in pure water. Kidney cortex and kidney medulla showed a difference in that the former became more swollen in acid and water than the latter. These experiments do not impress me as decisive because physiological salts were absent. The behavior of nerve tissue is especially interesting. Reichardt called attention to a clinical condition which occasionally occurs in dementia prsecox and causes sudden death. He noted an increase of weight and volume in such brains without other detectable macro- or METABOLISM AND THE DISTRIBUTION OF MATERIAL 231 microscopic changes. He designated this condition swelling of the brain in contrast with edema of the brain; it is possible to demonstrate the fluid transudate from the blood. In swelHng of the brain, it is dry, sohd and gelatinous. Reichardt's observation was made about the time Fischer published his theory so that it seems apropos to mention it. J. Bauer and Bauer and Ames tested slices of brain and cord and found that swelling was observed occasionally with a one thousandth normal acid- ity, but with greater concentration (even of carbonic acid), a shrinking always occurred. On this account they reject Fischer's acid theory. Contradicting J. Bauer, Fischer in experiments (with Hooker) found that nervous tissue behaved toward salts and acids Hke fibrin and other tissue. He explains the disagreement by the fact that Bauer chose tissue which had been dead 6 to 24 hours. The optimal concentration for sweUing had already been exceeded by the post mortem development of acid. The fourth objection was raised by pathologists who maintained that what Fischer described was not edema at all. The main loca- tion of edema, connective tissue, exhibits apparently, Hke fibrin, the swelfing and shrinking phenomena with acids and salts but is essen- tially dissimilar from edematous tissue and more like hyaline or amy- loid degeneration. (Marchand, Klemensiewicz, H. Schade.) The fibers show the chief swelhng in acids but in edema the main swelfing is extrafibrillar. Fischer fails to distinguish between swelling of the protoplasmic substance and turgor of the entire tissue. Against this view that in edema the aqueous fluids accumulate in the tissue spaces and not merely in the protoplasma and that increased inhibi- tion of water by the cell is not the criterion of edema, M. H. Fischer argues that the tissue spaces are not filled with air but by colloid ma- terial which may very well contribute to the edema by acid swelling. Lubarsch noted marked swelfing of the tissue in his histologi- cal studies but determined a difference in kidney edemas due to clamping the renal artery or renal vein. The changes are snxdlar; those produced by clamping the vein are reversible but those due to clamping the artery are irreversible if the artery is clamped off for three hours. The sensitive ceHs are kified. This contradicts Fischer's theory which requires the damage to be the same in either case. Kurt Ziegler made the üiiportant observation that chlorid metabofism as weU as water metabofism play an important part in edema. Chlorid and water retention alternate as primary factors in edema, but in all cases, there are nutritive disturbances which affect chiefly muscles and connective tissue. P. Tachau offered experimental verification by feeding mice excessive amounts of sodium salts. There occurred edema about the head, neck and attachment 232 COLLOIDS IN BIOLOGY AND MEDICINE of front limbs resembling that observed in human nurslings when they are given too much salt. In Tachau's experiments it is note- worthy that there was no increase in average amount of water in the animal but the edema was the expression of abnormal distribution of water. These observations lend some support to Fischer's theory. Fischer observed that when fibrhi and gelatin swelled up they accumulated salt from an acid table salt solution and, consequently, he regarded the salt retention as resulting from the edema. When we consider the controversy concerning Fischer's theory of edema, it is evident that as yet he has offered no experimental proof for his hypothesis. Except in a few special instances his opponents have likewise failed to show its invalidity since, in my opinion, their experimental methods failed to produce a local accumulation of acid in hving organs as required by Fischer's theory. No matter what value may be set on Fischer's theory in the future, it has been of enduring service in that it has transferred the emphasis in the study of edema from the circulation to the tissues; it is not hydrostatic differences in pressure but chemical damage to the tissue which occasions edema.^ Inflammation. Though healthy cells are impermeable for blood plasma, inflamed tissue permits a selective passage of plasma elements. A. Oswald * found that the frequency of passage stood in the fol- lowing order: Albumin > Globulin (Euglobulin) > Pseudoglobulin > Fibrinogen. This occurs in an order inversely to their susceptibility to salting out by salts and to the viscosity of the various solutions : the less viscous a plasma element is, the more easily does it pass through the inflamed tissue. Albumin alone may be found in an exudate, but never fibrin- ogen without the simultaneous presence of albumin and globulin. In the acute stages all kinds of albumins are found in the exudate, whereas, with the lapse of time, fibrinogen and then globulins diminish. The normal cell membrane evidently behaves Hke an impermeable ultrafilter, which has become more permeable by reason of the in- flammatory process. We do not know the factors which bring this about. [The tissue may be ''coagulated" allowing freer diffusion because of larger diffusion paths, or they may be more dispersed and accumulate more "water of swelling." Either condition would explain some of the phenomena. Tr.] ^ [The most recent discussion of the question by Lawrence J. Hendeeson and Martin H. Fischer is contained in the Journal of the Am. Chem. Society, Vol.' XL, No. 5 (May, 1918). Tr.] METABOLISM AND THE DISTRIBUTION OF MATERIAL 233 It is to be hoped that an extension of the viewpoint of M. H. Fischer, together with that of A. Oswald will explain the problem of inflammation, which has been so long kept at a dead center. We thus see that the science of colloids throws new light on the most difficult portions of pathology, and that even therapy may gaze with hope upon the young science. Salt Distribution. Just lis the water content of a normal organ is relatively constant and may undergo reversible changes due to changes in the condition of the organ colloids, this is also true in the case of the salt content. Unfortunately the basis for the comprehensive explanation of these questions is lacking (bibliography, see Albu-Neuberg *). The values that have been obtained cannot be used for comparison. Some have been derived from healthy and some (especially the de- terminations on man) from sick individuals; moreover, age is a very important factor. In the first place, it is essential to determine accurately the limits within which the salt content in the separate organs of normal individuals may vary. We know that muscle, liver, blood corpuscles and brain are rich in potassium salts, while in the blood serum and spleen, sodium salts predominate. In man the muscles contain 0.743 per cent of NaCl, the lungs, kid- neys and skin 2.5 per cent. Sodium chlorid and water content do not run parallel. The salt content varies within wide Hmits in different species of animals. For instance, there is present in the ash of Ox blood 7 . 4 per cent K2O Calf blood 11 . 2 per cent K2O Sheep blood 7. 1 per cent K2O Pig blood 20.4 per cent K2O Chicken blood 18 . 4 per cent KoO. We should assume from our colloid-chemical knowledge, that a given electrolyte content must correspond to each organ's condition of swelhng, though as yet this assumption offers but httle towards elucidating the problem. Pathological retention of common salt in edema, see page 232, invites experimental study of the question. What has been said of animals is also true of plants. But here, too, the salt content varies greatly with organ and sex, and we have no basis for its true significance. For instance, it is merely neces- sary to mention that the ash of wheat flour contains 0.76 per cent Na20, whereas that of buckwheat flour has 5.87 per cent Na20. The distribution of salts occurs similar to that of water; if they are artifically introduced into the organism, they are stored up and again released. If a dog received an intravenous injection of table salt, 28-77 per cent of the saline retained by the body accummulates in 234 COLLOIDS IN BIOLOGY AND MEDICINE the skin (Padtberg). In salt starvation the skin acts conversely, suffering 60-90 per cent of the chlorid loss. [The ancient name for eczema was "salt rheum"; see also " Karell Treatment," J. G. M. BuLLOWA, Amer. Medicine, June, 1918. Tr.] For animals, an intravenous injection of potassium salts acts as a poison (especially for cardiac muscle and peripheral vessels); thus, according to Held, even solutions with 0.08 per cent KCl affect frogs and rabbits. By way of the intestinal tract, potassium salts are relatively harmless. Held * showed that when thus in- troduced, the K was stored in the tissues and only slowly given up to the blood. Here we observe the same phenomena as with water, of which an excess is also taken up by the tissues. For the maintenance of the osmotic pressure, a definite concen- tration of any crystalloid suffices. Observations of the most differ- ent kind teach us that exactly that electrolyte which is normally found, is necessary for function and development. If in a suspension of blood corpuscles the NaCl is replaced by sugar, hemolysis becomes more difficult, even though the osmotic pressure is identical. Ex- periments on excised hearts prove that the action of the heart is re- tained much longer if it is perfused with a fluid containing a proper quantity of K, Ca, Mg, PO4, than if only physiological salt solution is used. Na, K, Mg and Ca salts are individually poisonous for plants, but mixed in the proper proportions they are absolutely necessary. In Chapter XXII, there are further examples. Evidently the condition of swelling required for normal function is afforded by a proper balance in the mixture of electrolytes. A.B. Macallum ^ investigated microchemically the distribution of K, Fe, Ca, CI and PO4 in many animal and plant cells, and from his investi- gation made deductions concerning their functional significance, to which we shall again return (see p. 292) . Th. Weevers has elaborated them with far reaching studies of the distribution of potassium in plant cells. [W. BuRRiDGE, Quarterly Journal of Medicine, 10, No. 39, p. 172, aptly remarks that analyses of the blood ash give little information con- cerning the balance of its salts by reason of the fact that the propor- tions of them which are in " sorption " or in solution may vary. Tr.] 1 A. B. Macallum proceeds in part from the fact that inorganic salts increase the surface tension of an aqueous solution, and as a result the surface contains a more dilute solution. Conversely the author concludes that the surface tension is diminished at the points of the cells which are approached by the salts in question. We cannot agree with this conclusion at present, because not only mechanical but chemical influences may determine the adsorption of salts. For instance, L. Michaelis and P. Rona (as the author has mentioned) demonstrated that certain kinds of sugar have no influence on the surface tension and yet may be adsorbed. METABOLISM AND THE DISTRIBUTION OF MATERIAL 235 There are, moreover, the interesting observations of P. Rona and D. Takahashi,* Hollinger* and E. Frank* concerning the dis- tribution of sugar between blood corpuscles and plasma. If we go further and observe the distribution of the other ele- ments of the organism, the albiunins, nucleins, elastin, the lipoids, etc., we approach the greatest problems of anatomy and histology and set out upon a boundless and uncharted sea. Possibly we shall learn more of these things in the not too distant future. The Circulation of Material. Both the plant and the animal organism are surrounded by mem- branes or pellicles, which separate them from the outer world. These membranes are more or less permeable to water and crystalloids, but normally, they are impermeable to colloids. Even though this last fact were not experimentally demonstrated, we should assume it a priori, because if the dissolved colloids could leave the organism, the loss of material would mean death. To demonstrate the correct- ness of our conclusion, we need but mention a pathological con- dition, albuminuria. In this condition the kidney becomes permeable for serum-albumin, and it is one of the physician's most important duties to compensate for the continuous loss of substance by proper dieting, so that no impoverishment of the tissues as regards al- bumin occurs. Within the organism also, there are many such partitions; they serve to organize activity, to guide the food along certain paths (arteries, veins, vascular bundles of plants) and to collect secretions (urinary bladder, gall bladder). The substances necessary to support Hfe must accordingly enter the organism as gases or crystalloids. In the case of plants, CO2 enters through the leaves; other foodstuffs, water and most of the inorganic salts (nitrates, phosphates, potassium and lime salts, etc.), enter through the roots. These substances are at the outset very diffusible and need no preparation. It is otherwise in the case of animals, which require outside of water but few crystalloids (sugar, salts) and are chiefly sustained by colloids (vegetables and meat). In order to enter the organism at all, these substances must first be changed to a crystalloidal condition. This is accom- plished by enzymes; the diastatic ferments split starches; pepsin and trypsin split protein; and herbivorse have ferments which are able to change even cellulose into a crystalloidal condition, etc. In like manner only gases or crystalloids can leave the organism (expired CO2, urine, perspiration). ^ 1 Feces, etc., do not, strictly speaking, leave the organism any more than diatoms which have been surrounded by an amoeba and then cast out (they are evacuated from a tube which passes through the animal). 236 COLLOIDS IN BIOLOGY AND MEDICINE The forces which accomphsh the entrance of food into the organ- ism and keep up the circulation of ^natter are in part purely mechani- cal, as performed by the lungs, the heart, the peristalsis of the intestines, etc. In addition to these, there are forces which ac- complish chiefly the metabolism of the cells; the most important of these are diffusion, osmotic pressure, swelling and shrinking.^ Circulation of Water. Until a few years ago the circulation of water in the organism was chiefly attributed to osmosis. The vital processes constantly produce from the colloids osmotically active crystalloids, which both retain the water formed by oxidation and, in addition, attract water into the cells, thus maintaining the turgor or normal tissue tension. This presupposes that an almost semipermeable membrane surrounds every cell. In the case of plants, this hypothesis offers certain difficulties, and in the case of animals, it is impossible to maintain it. We shall present only a few examples which show that osmotic conditions alone do not satisfactorily explain the distribution of water in animal cells. Through the investigations of H. J. Hamburger, H. Koeppe and E. Overton, it is known that in the presence of alterations of osmotic pressure, blood corpuscles and muscles change their volume to much less an extent than would be expected of cells with fluid contents and a semi-permeable membrane. Blood corpuscles contain about 60 per cent water. In his experiment with osmosis. Ham- burger showed that only from 40 to 50 per cent of their volume could consist of an aqueous solution, so that from 10 to 20 per cent of the water arises in some other way. According to Overton the same thing holds for frog's muscle. Water is also retained by swelling. Swelling and shrinking are the most powerful factors governing the circulation of water in the organism. They may even act against osmotic pressure; nor are we forced to explain their activity by any hypothetical membranes. Changes in the reaction of the cells, especially the constantly recog- nizable acid production during vital processes, give rise to the condi- tions necessary for swelling or the circulation of water. With the removal of the acids shrinking must occur again. M. H. Fischer properly cafls attention to the fact that a semi- permeable membrane permitting the entry and exit of water from the 1 J. Traube*! regards the "surface pressure" as the force which causes the movement of matter in the organism. Since there exists a certain parallelism between the ability of many substances to lower surface tension and their capacity to penetrate the cells, Traube disregards the osmotic forces and Lipoid solubility. METABOLISM AND THE DISTRIBUTION OF MATERIAL 237 cell is a monstrosity, for how does it explain the entry of food and the exit of metabolic products (metabolites) from the cell. If the mem- brane is permeable for these, the osmotically active crystalloids cannot induce transfer of water. All variations in volume of blood corpuscles, spermatozoa, plant cells, etc., produced by electrolytes and attributed to osmotic pressure up to now, are just as well ex- plained by swelling and shrinldng. Gels swell up in water, acids and alkalies; salts on the other hand, hinder sweUing and cause shrinkage. Moreover, new, purely physico-chemical observations likewise warn us to employ great caution in our consideration of osmotic processes in the organism. Much more substance may be dissolved in the in- terior of the cell than in its surrounding fluid without the osmotic pressure making this evident. We saw on pages 46 and 47 that with decrease in the surrounding osmotic pressure, the osmotic pressure in a cell with a permeable membrane containing colloid, falls, although if reckoned according to the salt content, the osmotic pressure should have increased. These findings of W. Biltz and A. VON Vegesack* necessitate a revision of all former conclusions derived from the observation of osmosis in cells. Circulation of Water in Animals. A movement of water results when conditions arise which change the relative swelling of the organs. When subjected to high tem- perature or after violent exercise, etc., the skin loses water, the blood loses water through the lungs, which causes a flow of water from the other organs. Conversely, an excess of water from the in- testines, or in the case of frogs and certain other animals from the skin, is transferred to other organs and re-excreted by the kidneys. Other circumstances may arise, however, which determine the cir- culation of water: concentration of acid in a tissue increases its swelling capacity, attracting water, e.g., in venous blood or an edema, whereas simultaneous salt formation leads to a shrinking or loss of water.^ In circumstances in which osmotic pressure may become active, as when a membrane is interposed, the change of a colloidal substance into a crystalloid under the influence of enzymes may effect a transfer of water; the water flows to the place where the osmotic pressure is higher. We shall return to the details of this question when we consider the individual organs. ' From this it results that the presence of colloids regulates the movement of water in an entirely different and at times in a direction opposite to that of the osmotic pressure: Acid + salt, as a result of the higher osmotic pressure, should increase the amount of water attracted; in the case of colloid structures, how- ever, they decrease it, since salts aboUsh to a greater or less extent the swelling action of acids. 238 COLLOIDS IN BIOLOGY AND MEDICINE The Movement of Water in Plants. The evaporation of water in plants proceeds more rapidly than in animals. The enormous development of surface in the shape of leaves and needles underlies a great transpiration which requires replacement, so that a stream of water moves upward through the roots and vascular bundles to the leaves. On bright summer days (see W. Pfeffer *2 loc. cit. I, p. 233) 1 to 10 gm. water are evaporated from 1 cm.2 of leaf surface. On very hot days the loss by transpiration from big trees exceeds 400 kilos; on rainy days, however, it may be reduced to a few kilos. To explain the upward movement of the water, the most varied theories have been advanced, and usually abandoned. Explanation by means of osmotic pressure has proved thoroughly unsatisfactory, and mere capillary imbibition is of no greater use, We may well understand that the colloids of leaves suffer a loss of water by evaporation, and that, in swelling, they are able to lift a great column of water from the ground to the tree top. Experiments of E. Strassburger showed that in poisoned trees, water may rise to a height of 22 meters, so that pure capillary forces do not suffice for the explanation of the phenomena. More recent ex- periments (P. A. RosHARDT,* E. Reinders*) show that in the living plant, living elements assist in pumping up the water. Since no pre- vious explanation of this has been given, I believe that I am justified in formulating the following hypothesis. In my opinion, the living cells of plants assist in the elevation of the sap by their respiration. With respiration, not only does CO2 develop, but also great quantities of organic acids. Both cause a swelling or attraction of water, which is liberated to the extent that CO2 disappears, and the other acids are removed in any one of the many possible ways. This would fit in with the fact that the breathing in fully developed leaves and branches, in which the need for water is also diminished, is less than in the developing shoots. The dead leaf, whose breathing has ceased, withers. Circulation of Crystalloids. The circulation of crystalloids is also largely governed by the factors of diffusion and osmotic pressure, with certain limitations due to the colloid media. Although between two aqueous solutions, sep- arated by an easily permeable membrane, unrestricted mixing occurs as a result of diffusion, this does not hold for a jelly-like medium (see H. Bechhold and J. Ziegler*^). In order to bring about a mixture in such a case an excess of osmotic pressure is required (see p. 57). It even seems that with equal osmotic pressure, acid METABOLISM AND THE DISTRIBUTION OF MATERIAL 239 and alkaline reacting substances may lie side by side in colloidal (amphoteric) media for a long time without neutralizing each other (R. E. Liesegang). In the case of phagocytes, that is, in Hving cells, the existence of acid areas in alkaline protoplasm has been shown by staining with neutral red (E. Metschnikoff). We thus see that in different portions of the organism, the most various crystalloids are present, and may functionate specifically without being accompanied by any exchange or mixture; this only occurs when a crystalloid sub- stance accumulates and becomes osmotically active. Swelling and shrinking may also be of importance for the circulation of crystal- loids, since dissolved substances are soaked up with the water of swelling or are expressed during shrinking. If these crystalloids are at the same time electrolytes, they may increase or diminish the swelling according to their nature (acid or salt); and in this way, either aid or impede the entrance of crystal- loids. Circulation of Colloids. Compared with crystalloids, the osmotic pressures in the case of colloids are extremely small. To be sure, we know (see p. 55) that proteins may diffuse through gels, so that they also are independ- ently motile. Of great significance is the discovery of H. Iscovesco to the effect that colloid diffusion is dependent on the electric charge. In general, however, the colloids, as opposed to the crystalloids, furnish the stable element of the organism. The Influence of Membranes Upon the Interchange of Substances. The physico-chemical conditions for the interchange of substances through cell membranes was for a long time completely ruled by the theory of Overton, which is somewhat as follows: Protoplasm is surrounded by a fatty lipoid membrane; an exchange of substances can only occur if the given substance is soluble in such a membrane. Overton's theory has not proven universally applicable; it is ever becoming better recognized that the problem will probably be solved when we cease to look entirely to the osmotic conditions and mem- branes for the factors governing the interchange of substance. The fact that both cell content and cell membrane consist of colloids capable of swelling must be taken into consideration. The earliest fundamental investigations of the physical inter- change of matter in individual cells were made on plant cells. I refer particularly to the investigations of W. Pfeffer and H. de Vries. In plants, especially, we find that the cell content is very fre- 240 COLLOIDS IN BIOLOGY AND MEDICINE quently surrounded by a visible and solid membrane which is usually regarded as semipermeable. The basis of this view is: if such a cell is placed in hypertonic salt solution, the protoplasm retracts from the cell wall and water is lost. This phenomena is called plasmolysis. If the cell is im- mersed in pure water, the protoplasm swells up again. The phe- nomenon was formerly explained by saying that the membrane was impermeable for salts.^ In a hypertonic salt solution, water may in- deed leave the cell but salt cannot enter; in pure water the process is reversed. Nowadays discussion is focussed on the nature of the plasma pelHcle and two main tendencies may be recognized. Among the adherents to the lipoid theory in addition to E. Oveeton, is Vernon, who considers it probable that the hpoid membrane penetrates the interior of the cell. J. Loeb and R. Beutner in view of their investigations of bio-electric phenomena may be regarded as adherents of the Hpoid theory. Those investigators (J. Traube and F. Czapek) who regard changes in surface tension as the means of penetrating surfaces may be regarded as adherents of a modified lipoid membrane theory. They arrive at this conclusion because their experiments have been chiefly concerned with the action of lipoid soluble substances on the cell. According to F. Czapek all substances whose surface tension is less than 0.68 (water/air = 1) are toxic for the higher plant cells and Czapek's pupil KiscH determined 0.5 to be the hmit of toxic surface tension for yeast cells and fungi. Since lecithin and Cholesterin, that is, the lipoids and their emulsions, have a surface tension of 0.5, F. Czapek agrees with Nathanson and regards the cell membrane as a concentrated fat-emulsion which is permeable for either fat or for water soluble substance depending on the conditions of surface ten- sion. Similar views (loose union of albumin and lipoid) are enter- tained by W. W. Lepeschkin with the difference that he regards the entire protoplasm as such an emulsion possessing properties in the center similar to those on the surfaces. The "emulsion theory" obtained very definite support from the fol- lowing observation of Clowes (see p. 38). He prepared an oil-water emulsion by shaking equal quantities of water and olive oil and suffi- cient n/10 NaOH that the outer phase (the water) was just alkahne to Phenolphthalein, If he now added a small excess of CaCl2 solution the emulsion changed into a water-oil emulsion; in other words, water became the dispersed phase in a continuous layer of oil. We 1 The visible cell membrane is quite permeable for most crystalloids, serving only to a certain extent as a support for the protoplasm. METABOLISM AND THE DISTRIBUTION OF MATERIAL 241 observe that by this chemical attack the layer which had been per- meable for hydrophile substance became impermeable for them and was made permeable for substances soluble in fat. However, we know from the investigations of J. Loeb and W. J. V. Osterhout (see p. 378, et seq.) that small amounts of divalent cations detoxicate neutral salts by inhibiting, according to the view of J. Loeb, the free ex- change of ions through the plasma pellicle. Clowes extended his observations to other polyvalent cations and the quantitative rela- tions are in excellent agreement. The other tendency is to assume a pure albuminous membrane; W. J. V. Osterhout assumes this, as the result of the following remarkable observation: he placed spyrogyra cells in common salt solution of such concentration that no plasmolysis^ occurred; when he added very dilute calcium chlorid solution so as to depress the osmotic pressure, plasmolysis occurred. The plasma pellicle must have been permeable for NaCl and its passage is only impeded by the CaCl2. In contrast to Overton's view, Osterhout regards the plasma pellicle as permeable for most ions of the light metals and consequently it must be albuminous. The action of the Ca-ion possibly depends on its antagonistic action (see p. 69) though it may be due to a variety of tannage of the plasma pellicle. With the death of the cell, the pellicle becomes generally permeable. RuHLAND also, discards the lipoid theory. He considers only the thickness of the membrane to be responsible for permeability or impermeability; the membrane acts like an ultrafilter in the sense of Bechhold. He studied a large number of dyes, enzymes, alkaloids and other substances which occur in plants and found that their ability to penetrate the plasma cells was in proportion to their ability to spread out in thick jellies; in other words, it depended on their particle size (see p. 56). In view of the known facts we must admit that at present we can arrive at no conclusion concerning the nature and structure of the plasma pellicle. Of one thing we can be certain, that Overton's original theory of a continuous lipoid membrane must be abandoned. I am of the opinion, however, that it is possible to conciliate the theories which have been elucidated here and which seem to be mutually exclusive. In the first place, the assumption of a pellicle of emulsified fat does ^ Osterhout distinguished between true and false plasmolysis. The latter may occur in dilute solutions even in pure water most usually in marine plants. It is probably due to the coagulation of the protoplasm from the penetration of the water. 242 COLLOIDS IN BIOLOGY AND MEDICINE not exclude Ruhland's ultrafilter theory. If we have an emulsion in which the lipoid is the dispersed phase we have an ultra filter which is permeable for water soluble substances and impermeable for lipoid soluble substances. The size of the pores of the ultra filter depends on the relation of the lipoid to the aqueous phase. If the amount of lipoid is small, the pores of the ultra filter are large and vice versa. When the lipoid content is large we have a narrow pored ultra filter which absolutely satisfies the conditions found by Ruhland in his dye investigations. Further, we have seen from Clowes' experiment, that an oil/water emulsion is easily changed to a water/oil emulsion, and in that case the layer is open for fat soluble substances and closed for water soluble substances In my opinion such a layer satisfies all the conditions demanded by the various investigators. I wish, however, to emphasize that there is no justification for too wide a generalization, for different cells behave very differently. Observations on plant cells cannot be applied without modification to animal cells; a cell in a plant root cannot be compared to nerve cells which are surrounded by a dense isolating layer of fat. It seems possible to conclude from R. Höber's and Ruhland's experiments on the penetration of dyes into cells that the animal cells which they studied contain larger pores than the plant cells. Let us consider the simplest instance, one in which the cell proto- plasm is a colloid capable of swelling, with surfaces limited by a pellicle which can also swell and offering certain exterior boundaries. Any injury to this pellicle will be repaired of its own accord somewhat like rubber. We can thus (pp. 284-285) readily understand how amoeboe or phagocytes send out protoplasmal prolongations, envelop foreign bodies or bacteria and incorporate them without their margin being broken. As a matter of fact, it must immediately repair itself just as does an oily film on water broken by a stone. Let us see how this view agrees with former theories, and to what extent this view is an improvement upon them. To begin with, it must be noted that Overton assumes that a sub- stance is taken up by the plasma pellicle in accordance with its coefficient of solubility, in agreement with the laws of solutions (Henry's distribution). This may be the fact in many cases, only we must recall that adsorption fulfills similar conditions for the passage of a substance through the plasma pellicle into the interior of the cell. The only condition which need be assumed in order that a substance may enter the interior of a cell from outside, is that there shall be a reversible absorption by the plasma film. What curve of distribution this follows, is immaterial for the present. That, as a matter of fact, in numerous cases an adsorption certainly does exist, METABOLISM AND THE DISTRIBUTION OF MATERIAL 243 but not a distribution according to Henry's law, has been determined by H. Bechhold in the action of disinfectants (see p. 399), and Straub-Freundlich on the distribution of veratrin between heart muscle and pericardial blood. G. Loewe has shown by simple physico-chemical experiments that lipoids adsorb dyes, narcotics, nicotin and tetanus toxin. The substance interchange in animal cells has been studied most thoroughly in the case of red blood corpuscles. In my opinion (see p. 304) the latter have a very peculiar struc- ture, conditioned by their special function; their lipoid pellicle is quite strong. In spite of this, we shall find phenomena in the case of the erythrocytes which cannot be brought into accord with the idea of a salt solution surrounded by a semipermeable membrane. The theory of osmotic pressure demands that various isotonic salt solutions shall have equal influence upon the volume of the blood corpuscles. S. G. Hedin*^ showed, however, that this is not the case, for instance, in isotonic solutions of NaCl and KNO3; in the case of lower concentrations, the volume is smaller; in the case of higher concentrations it is larger than with the corresponding NaCl solu- tion; we must recall that the NO3 ion favors swelling or the de- flocculation of colloids and lecithin; if the outer pressure is low, crystalloids leave the blood corpuscles, and the osmotic pressure, and consequently the volume of the corpuscles, will be less than with the corresponding NaCl solution. The reverse occurs if the outer solu- tion is hypertonic. We find in the literature, repeated references to the permeability of the cell membrane, especially of plants, for 'potassium nitrate. B. VAN Rysselberghe* has demonstrated the entrance of diphenylamin into tradescantia cells. If fungi, such as aspergillus niger or penicil- lium glaucum, are grown upon a concentrated solution of saltpeter, they will take up so much of the electrolyte that in the end they will have an osmotic pressure of 200 atmospheres. Such cultures actually explode when placed in pure water. The ability to take up such substances as favor swelling is much greater in the case of young cells with membranes that can swell than in the case of old inelastic cells. On this account, an older aspergillus cell may plasmolyze with a 20 per cent NaNOs solu- tion which possesses an osmotic pressure of only 102 atmospheres. In this difference between old and young cell membranes, may lie a partial explanation why bacteria and fungus cultures, namely organ- isms which multiply rapidly, readily adapt themselves to changed conditions. Young and old cells differ in their turgidity. R. HöBER*^ prepared suspensions of blood corpuscles in dilute iso- 244 COLLOIDS IN BIOLOGY AND MEDICINE tonic solutions of various alkali salts, and observed the order in which they favored hemolysis. He established the following series: S04< CI < Br, NO3 < I, and Li, Na < Cs < Rb < K. The anion series corresponds fairly well with the action of the anions upon lecithin, so that we may safely assume that alkali salts may bring about an increase or a diminution in the porosity of the plasma pellicle. Other examples of this action on the part of neutral salts are given in Chapter XXII (Salts). Substances which favor swelling in a diluted condition (p. 68) may prevent it when they are more concentrated. It must also be borne in mind, that besides hydrophile lecithin and albumin, hydro- phobe Cholesterin must also exist in the lipoid membrane. This lat- ter is, however, precipitated by electrolytes, which cause the former to swell. Thus there exists in the cell membrane a self -regulating system, something like a compensation pendulum; when the tem- perature rises, the center of gravity of the pendulum falls, but by a combination of metalhc rods the center of gravity is raised and the fall is compensated. This compensatory action of hydrophile and hydrophobe colloids appears to be an essential factor in the automatic regulation of cell metabolism. The following considerations afford an explanation for some par- ticular kinds of cells. R. Höber*^ properly calls attention to the fact that, ''the plasma film is really impermeable for everything the cell needs or produces." It is impermeable for amino acids, for the various kinds of sugar and soluble carbohydrates which are formed in the interior of a cell from the undissolved carbohydrate reserves and for inorganic salts and salts of organic acids. The in- terchange of these substances, which naturally rnust occur, is on this account, somewhat of a riddle. In my opinion, these phenomena are less mysterious if we recall that with equal osmotic pressure without and within, even the thinnest membranes interfere with diffusion, as has been shown by H. Bechhold and J. Ziegler (see p. 57). The most recent investigations of H. J. Hamburger on blood corpuscles indicate that their transition membrane is not as impermeable as was formerly believed. It will be understood thus how the cells are sharply cut off in case of isotonicity, while if there be hypertonicity, some substance may penetrate through the mem- brane. This seems to me to be the meaning of the following ex- periment of J. Bang* : if red blood corpuscles are placed in an 8 per cent cane sugar solution and the solution is immediately diluted, hemolysis occurs when the cane sugar concentration is 5.4 per cent. If, however, the red blood corpuscles remain in the cane sugar solu- METABOLISM AND THE DISTRIBUTION OF MATERIAL 245 tion for several hours and the dilution is then undertaken, the hemolysis will occur only when the cane sugar concentration reaches 2 per cent. Sufficient time has thus been given for salts to leave the blood corpuscles and enter the cane sugar as was shown by A. GiJRBER. Experiments of Jaques Loeb*'* upon the parthenogenesis of sea urchin's eggs are in accord with this. If the eggs are placed for a short time in hypertonic salt solution and then returned to sea water (which corresponds with their normal osmotic pressure) seg- mentation takes place. J. Loeb** found that a cane sugar solution acts like a hypertonic salt solution even if, as regards concentration, it be isotonic with the eggs. J. Loeb explains the action by saying that the egg pellicle is permeable for sugar and salts, and that the salts diffuse out more rapidly than the cane sugar diffuses in, so that the outer fluid becomes hypertonic. We must, moreover, recall that substances exist which to a certain extent close the pathways auto- matically, as for instance the SO4 ion, whereas others, especially urea, open a passage, not only for themselves but for other substances (see p. 55). The permeability of red blood corpuscles and muscles for urea is then no longer surprising, any more than the changes in permeabihty (observed by M. Fluri* and R. Meurer*) in the plasma pellicle of plants under the influence of certain salts. This may be accomplished not only by chemical agencies, but purely physical factors may have an influence. It might be ex- pected a priori from change in temperature; the influence of light is surprising, as experiments of W. W. Lepeschkin and by A. Tröndle have shown. The latter's experiments indicate that plant cells (foliage) are more permeable, not only for NaCl but even for glucose, in a bright light than in the dark. One of the most remarkable and still unexplained phenomena is that when death occurs, the permeability of the cell membrane changes into that of an ordinary membrane which retains only col- loids. Assimilation and Dissimilation. After a crystalloid foodstuff has entered the organism, it is the organism's most important task to retain it for use; this is ac- complished by changing it into a colloid, inasmuch as complicated combinations are formed from more or less simply constructed crys- talloids. From the CO2, which enters the leaf, starch is formed under the influence of chlorophyl granules and daylight; and from nitrates which have entered through the roots, with the assistance of carbohydrates, proteins develop. In the animal organism, the 246 COLLOIDS IN BIOLOGY AND MEDICINE readily diffusible peptones which have entered through the intestine are again changed even in the intestinal wall into colloidal albumin; sugar is retained in the liver as an animal starch or glycogen and so on for other examples. This change into the colloidal condition is usually associated with a metamorphosis into a substance native to the body from which the cells and tissues of the organism are built up. It may be con- cluded from the investigations of Alexis Carrel and Burrows* that the circulating nutritive fluid already contains all the elements required for the most varied organs. These investigators suspended pieces of tissue from freshly killed mammals in drops of plasma from the same kind of animal. The tissues continued to grow, cartilage produced cartilage, a spleen produced cells which resembled spleen pulp, and pieces of kidney grew tubes of cells which resembled the kidney tubules. The future will teach us whether we must regard this phenomenon as a kind of crystallization. Perhaps it will be possible for future colloid investigators to express the problem of cell nutrition in terms of the brilliant side-chain theory of Ehrlich. To what extent the fixation of colloid foodstuffs by the cell is a matter of chemical forces or simple mechanical adsorption is an open question, which so far must be decided differently in each individual case. Hitherto it has been only possible in the case of fats, to follow visibly their course from the moment of resorption to their fixation in body tissues. In the intestines^ with the assistance of the alkaline reaction of the intestinal fluids, the intestinal and pan- creatic juices and the bile form a very fine emulsion of the fats. Simultaneously, there occurs a splitting into fatty acids and glycerin under the influence of ferments, lipases. It is not yet established whether the splitting of the fats is complete, which would mean that the intestine could only absorb dissolved fatty acid salts of alkalis (soaps) and glycerin, or whether some of the fat remains unchanged and is absorbed as such. If this latter statement is actually true, we should have to assume that, under the influence of the surrounding soap solution and possibly other factors, the surface tension of the fat droplets is reduced to a minimum so that they may easily change its shape or enlarge their surface and pass the very minute openings in the intestinal epithelium. From page 16, we know that to produce a change of form by pressure alone, there would be required forces (many atmospheres) such as never occur in the organism. We have in this case conditions similar to those in the case of leucocytes (see 1 It is hardly possible to attribute great significance to saponification in the stomach. METABOLISM AND THE DISTRIBUTION OF MATERIAL 247 pp. 283 to 286), which, in spite of their considerable diameter, change their shape so that they pass through the finest vessel walls. If to the leucocytes is to be attributed a share in the resorption of fat, they must journey into the intestines and return laden with fat. It has not been possible as yet to decide microchemically whether fat passes the intestinal epithehum unchanged. [In the presence of protective colloids, colloidal gold will pass through Pukall filters which other- wise hold them back. Zsigmondy-Alexander, Colloids and the Ultramicroscope, p. 153, et seq. Tr.j One fact, at least to me, seems very much to favor the idea that fat may be resorbed unchanged from the intestines, namely, linseed oil, sesame oil, cottonseed oil, etc., may occur unchanged in the milk, and foreign fats (rapeseed oil) may be deposited in the body. Absorption occurs almost exclusively in the small intestines (Naka- shima). Within the intestinal wall, neutral fat may be synthesized from the absorbed fatty acid alkali and glycerin, so that neutral fat is carried to the body in very fine emulsion through the chyle ducts, and in fact fat may enter the blood stream directly. The milky turbid lymph collects in the thoracic duct and empties into the subclavian vein. It is especially easy after the ingestion of fat to recognize ultramicroscopically, in the blood, numerous gran- ules (hemoconia), which may be considered fat droplets (A. Neu- mann,* K. Reicher*). It is possible, therefore, to follow visually the path of fat by means of dark field illumination. S. Bondi and A. Neumann* experimented as follows : at times, they caused hemoconia to appear in the blood by a liberal fat diet, and at others they injected a very fine emulsion of fat into the veins. Large fat droplets suspended in blood are evi- dently unable to change their form, and as a result cause emboli in the lungs, which may prove fatal. These investigators experimented thus with emulsions of lanolin, Cholesterin, lecithin, butter and olive oil, whose particles were only recognizable in the dark field. They dissolved the fat they were using in alcohol and poured the alcoholic solution slowly into water with constant stirring. The alcohol was removed from the filtrate of this emulsion by gently warming it on a water bath. S. Bondi and A. Neumann then established that the fat droplets were not dissolved by lipolytic ferments during their sojourn in the blood.^ They are emulsified by the venous blood in the right heart, and after they have passed the capillaries of the lesser circulation, they enter those of the greater circulation. Their goal, like that of ^ In my opinion such emulsions of uniform particle size could serve in measur- ing the exact dimensions of the smallest capillaries under normal conditions. 248 COLLOIDS IN BIOLOGY AND MEDICINE other suspensions (India ink, collargol) is the liver, spleen and bone marrow. This was demonstrated by intravenous injection of stained fat suspensions (lanolin with indophenol — fat with scarlet R).^ In these organs there are certain ceUs which take up the fat par- ticles (in the liver the star cells of von Kupfer). It should be again emphasized, that the fat particles behave exactly like inorganic suspensions. They behave like inorganic suspensions, also as regards rapidity of deposition; (depending on the size of the animal) after an emulsion is poured into the blood (by alimentation or injection) one- half hour to an hour suffices for it to disappear from the circulation. It follows from the above that in the deposition of fat there is no specific kind of fixation; that there is no solution but a purely me- chanical retention of fat particles in these storehouses. How the storage of fat occurs in other organs, and how the mobili- zation of fat reserves must be pictured, whether as an exceptionally fine emulsion or a true solution, all these are still open questions. An especially instructive example of research regarding an assimila- tion process is that of WOOD FORMATION. From a chemical standpoint, the lignified cells of the plant (wood) consist of protoplasm, cellulose and lignin. While cellulose may be considered as a distinct substance, a highly polymerized carbohydrate, fittle has been known regarding the constitution of "lignin." Evi- dently it is a mixture of various vegetable gums, pectins, lignic acid, albumins, glucosides, tannins, vegetable coloring matter, resins and other incidental constituents. A substance in lignin which is re- garded as characteristic and which stains with anilin salts and phloroglucin hydrochlorid was isolated by F. Czapek and identified as an aromatic aldehyd. Various theories have been proposed to account for the formation of wood; some placing more stress on physiological cnanges, and others attempting to explain the process on a purely chemical basis. I can dismiss these investigations in view of the fact that H. Wisli- CENUS* has offered and experimentally established a theory which places the entire view of this question as well as its experimented investigation upon a new basis. He assumes that the cambial juice which penetrates the cambial tissue stored between the wood and the inner bark layer during the summer vegetative activity, contains crystalloids (salts, sugars and plant acids) as well as colloids. These same colloidal constituents (formative substance or procambium) are all found in the lignin. According to H. Wislicenus, the process of wood formation occurs in three stages. ^ For further references see S. Bondi and A. Neumann loc dt. METABOLISM AND THE DISTRIBUTION OF MATERIAL 249 1. Formation of cellulose hydrogel in the youngest plant tissues as a chemically indifferent surface or framework. This primary stage will be explained more fully. 2. The colloidal constituents of cambial juice become layered upon the cellulose surface by adsorption and gel-formation, and thus in- crease the thickness of the surfaces. 3. Chemical reactions occur between the adsorbed hydrogels which lead to lignin formation. The following facts indicate the truth of these assumptions: (a) Colloidal constituents can be extracted by adsorption from the cambial juice. (6) The colloids adsorbable from the cambial juice and the run- ning sap are indicative of their hgnin content, which means their wood-forming properties. (c) The quantity of adsorbable colloidal constituents in the cambial juice varies with the season of the year — (shown in the annular rings as early summer and late summer wood). Certain trees give such large quantities of spring sap on tapping, that it is easy at times to collect a liter or more in a day. In North America the sugar maple has this property. In Norway, Sweden and Russia people drink the sap of the birch either fresh or fermented. The trees are bored about 10 cm. deep, 30 to 40 cm. above the ground and a glasp tube is inserted and sealed in with tree-wax. The sap drops through the bent tube into a bottle. To obtain the cambial sap, trunks of birch, pine and gray ash are sawed into pieces, 15 to 20 cm. long. The bark from 7 to 15 kilos of this is taken and then spht vertically. The smooth inner layer of the bark and the outer cambial mass of the smooth surface of wood are well shaved off with glass. The shavings are placed in from 1 to 2 liters of water and allowed to remain several hours, a few drops of thymol solution being added. The water is poured off, the residue squeezed in a fruit press and the combined turbid fluids are filtered. The adsorption experiments were performed partly by shaking with finely divided cellulose (filter paper) and partly by siphoning through "washed clay." (See p. 110.) In both cases the quantity of material adsorbed was estimated by determining the weight of the dried residue of (a) a measured quantity of fluid before adsorp- tion, (6) the same volume after adsorption. H. WiSLicENUS was able to prove in the case of the rising sap obtained by tapping the hornbeam and in the cambial juice of the birch, that the abstraction of colloidal substances followed an adsorp- tion curve, because the more dilute the solutions the more eolloida' 250 COLLOIDS IN BIOLOGY AND MEDICINE substance relatively was extracted from them by "fibrous clay" or filter paper. It appeared furthermore, that the rising sap of the birch contains only a small quantity, from 3.5 to 8.4 per cent of dry colloid substance. From this fact Wislicenus concluded that in the rising sap (until about the opening of the leaves — end of April) there is no new for- mation of colloids, but there occurs a dissolving of everything that is soluble (partial reversal of wood formation). The cambial juice on the other hand contains, at the time of most active wood formation (end of May to end of July) large quantities of adsorbable colloids (24 to 37 per cent). Towards the end of July or beginning of August, when wood formation quickly ceases, the colloid content of cambial sap also decreases rapidly. p. ( beginning of July 31.1 per cent adsorbed colloids. ( beginning of August 6.41 per cent " " p , ( beginning of July 24. 19 per cent " " ( beginning of August 8.04 per cent " " Thus H. Wislicenus has convincingly demonstrated that wood formation is a colloid-chemical process. Enzymes without doubt play a most important part in the de- velopment of organs. We know that enzymes serve the body by changing colloids into crystalloids. They split albumin into Poly- peptids and amino acids, starch into saccharids, etc. Construction or synthesis is also brought about by enzymes. Reactions which are hastened by enzymes are reversible and it depends entirely upon sur- rounding conditions, whether the balance of the process weighs more in one direction than in another. Thus, for instance, A. Croft Hill was the first to show that the same ferment which splits maltose into glucose actually forms maltose in a concentrated solution of glucose. Since then, similar reversals have been frequently observed: Potte viN split fats by means of pancreatic lipase and with the same enzyme he also prepared fats from glycerin and oleic acid. The well- known cleavage of fats with the enzyme of the castor bean was so successfully reversed by Welter that he obtained synthetically almost 30 per cent of neutral fat with the same enzyme. In general the cleavage process proceeds best in the presence of much water, whereas synthesis is most favored by the absence of water. By the swelling or shrinking of the colloids present during the reaction, the organism is able to permit the process to proceed in one or the other direction. Swelling and shrinking are in turn de- METABOLISM AND THE DISTRIBUTION OF MATERIAL 251 pendent upon the formation and removal of acids by oxidative proc- esses. The concentration of the products of a reaction in the solution brings the reaction to a standstill. If we are dealing with cleavages the crystalloid products may be readily removed by diffusion. This does not occur m the case of synthetic colloid products. Nor is it so essential because from the point of view of the law of mass action they are not to be regarded as dissolved. We know enzymes to be the excreta of the stomach and intestines, but we also know that cells themselves contain enzymes; we may mention the uricolytic enzyme of the hver and zymase of yeast which ferments sugar. We must also recall that enzymes are the most strongly adsorbed of all the colloids of the body, and that the ability to be adsorbed is largely dependent on the acid or alkaline character of the medium. An enzyme may be so fixed (e.g., rennet by charcoal) that its very existence is no longer determinable; a change of reaction recalls it to life. It may also be released by the approach of another colloid (in our example casein) by which it is adsorbed still more strongly. We thus get an inkling of the great importance enzymes have for the life of the cell, without as yet understanding the details. The conditions governing dissimilation are much more readily un- derstood. Through enzymatic cleavage of colloids, there are formed crystalloid products which pass into the circulating fluids of the or- ganism by diffusion and leave as excreta, or, after oxidation to CO2, are expired. We must not conclude that in every instance the entire colloid molecule breaks down into crystalloid cleavage products. In this way, by the splitting off of individual "side chains" (P. Ehrlich) there is permitted great variation in cell fife, which we might assume from our previous experiences. CHAPTER XV. GROWTH, METAMORPHOSIS AND DEVELOPMENT. Growth. Of all the problems in biology, one of the most difficult and most engrossing is the development to constant type. From cells, which externally can hardly be distinguished, we see develop a quail, an oak tree, a butterfly or a man. In their evolution they always pass through the same stages to the same ultimate forms which after a progressive senescence, return to the eternal process of evolution. If we try to reduce these developmental processes to their sim- plest terms, we find diffusion and swelling phenomena with the formation of precipitate-membranes. E. F. Runge, who discovered carbohc acid in coal tar, and who made the first anilin color,^ published a book in 1855, which is one of the most original scientific diversions I have ever seen. It is called "Der Bildungstrieb der Stoffe'' (The formative instinct of matter) viewed in automatically developed figures By Dr. F. E. Runge. (Oranienburg. Printed by the Author.) The book consists of a collection of blotting paper leaves, upon which various inorganic salt solutions were dabbed; they interacted and gave colors by which the most remarkable figures were produced. At first glance these seem to be lower forms of animal life, amebse or rhizopodse, and the collection, just as Häckel's ''Kunstformen der Natur," might well serve as a text for designers, because it offers such a multitude of suggestions with respect to color and shape. All the pages of the collection were prepared by the author himself (not printed) and are accompanied by a small amount of text which explains the method of preparation. The explanation of these creations is easy to the author, who says in one of his conclusions: ^ [Chas. Lowe is regarded as discoverer of phenol by the English, and Sir Wm. Henry Perkin is generally acknowledged, even by Germans, to be the discoverer of the first anilin color, mauve. Tr.] 252 GROWTH, METAMORPHOSIS AND DEVELOPMENT 253 "After all, I believe I may make the assertion that the creation of these pictures is due to a new and hitherto unrecognized force. It has nothing in common with magnetism, electricity or galvanism. It is not stimulated or created by any external force but is innate in substances and becomes active when their chemical affinities neutralize themselves ; that is, they undergo selective attractions or repulsions, and thus combine or sepa- rate. I call this force ' Formative Instinct ' and regard it as the prototype of the 'vital force' of plants and animals." Of course, no special "Force" need be invoked for the explanation of Runge's pictures. They are the result of very complicated diffusion and capillary phenomena associated with chemical trans- formations. What is especially interesting in Runge's pictures, on the one hand, is the constancy of the forms obtained by employing similar substances, and on the other hand, the extraordinary multiplicity brought about by the diverse action of different substances. If we let a drop of copper sulphate solution fall on a piece of filter paper moistened with potassium hydrate, at the surfaces of contact a membrane of copper hydroxid forms, which changes rapidly but always in the same way. If we always employ the potas- sium hydrate and copper sulphate in the same concentration, the copper hydroxid bomidary will always have the identical form, provided the same filter paper is used. A change in the concen- tration of one or the other ingredients, however, gives a membrane of different shape. If, instead of copper sulphate, we place a drop of copper nitrate on the paper, we obtain forms entirely different, and a drop of nickel sulphate changes the picture completely. We thus see that small variations in the concentration of the solutions and in their chemical composition possess numerous possibilities for the formation of new shapes. In the living organism variations in concentration perpetually occur. We know from biological reactions that not only different animals, as, for instance, sheep and lions, have chernically different tissues, but that even§the ass and the horse, and indeed different races of men, may be chemically differentiated; consequently the second condition for variation in form is also given, namely, the difference in chemical composition. The processes of the body (organism) are regulated by its colloidal state, and this very colloidal state also permits the reteji- tion of shapes. If we seek to leave this far too general point of view and study details of the question more closely, we encounter almost insur- mountable difficulties. If a small lump of copper sulphate is thro^vn into a dilute solution 254 COLLOIDS IN BIOLOGY AND- MEDICINE of potassium ferrocyanid, there will soon develop a brown envelope which throws out upward-growing runners, and in half an hour's time the fluid is filled with figures which vividly recall both the shape and the color of seaweed; if a small amount (0.5 per cent) of gelatin has been added to the water, the figures have some stability. Their development is easily explained: the copper sulphate dis- solves and immediately forms a semipermeable membrane of copper ferrocyanid, through which no copper sulphate can escape, but water may enter. Since a concentrated solution of copper sulphate is formed within, water will be absorbed until the membrane bursts^ whereupon the copper sulphate solution is brought into contact with the potassium ferrocyanid again and forms a new pellicle of copper ferrocyanid, and thus the process goes on. Stephane Leduc has studied these figures most industriously and has discussed their significance in numerous publications.^ Some of his directions are here given: Prepare granules of 1 part sugar and 1 or 2 parts copper sulphate. This is scattered in a fluid consisting of 100 parts of water and from 10 to 20 parts 10 per cent gelatin, 5 to 10 parts saturated potassium ferrocyanid solution and 5 to 10 parts saturated sodium chlorid solution, which mixture has been heated to 40 degrees Centigrade. In this way we obtain figures Uke that in Fig. 41, which may attain a height of 40 cm. The gelatin is solidified by cooling and the figures may be preserved. Other fig- ures are obtained by throwing granules of fused calcium chlorid or barium chlorid into a concentrated solution of soda. Another recipe is: Water 1 liter 33 per cent potassium water glass solution 60 gm. Saturated soda solution 60 gm. Saturated sodium phosphate solution 60 gm. Beautiful branching figures are given by scattering calcium chlorid granules in this mixture. The more concentrated the solutions, the more rapid is the growth and the more branching and delicate are the shapes. If the outer water is diluted while the growth is in progress, we may produce figures with stems and tops, like fungi (mushrooms and toadstools, etc.). These figures react to small changes in osmotic pressures by changing their shape. ^ I shall mention only his latest pubhcations: St. Leduc, Biochem. Ztschr. (Festband f. H. J. Hamburger), 1908, 280 u. ff. — Les croissances osmotiques et I'origine des etres vivants (Bar-le-Duc, 1909). Les bases physiques de la vie et la biogenese (Presse Medicale, 7, 12, 1909). Theorie physico-chimique de la vie (Paris, 1910), La dynamique de la vie (A. Poinat, Paris, 1913) ; and many others. GROWTH, METAMORPHOSIS AND DEVELOPMENT 255 St. Leduc calculated that a granule of potassium ferrocyanid acquired 150 times its original weight when it grew, and that calcium structures acquired many hundred times their original weight. The internal structure, also, has some similarity to natural forms. They have a cellular structure as St. Leduc showed by microphoto- FiG. 41. Artificially prepared osmotic seaweed. (Made by St. Leduc.) graphs, Fig. 42, and as could be inferred from the way in which they are formed. The solution, e.g., of copper sulphate, which is sur- rounded by a membrane of copper ferrocyanid, imbibes water until the membrane bursts, and some copper sulphate solution is exuded, which straightway surrounds itself with a film of copper ferrocyanid, and thus a new cell is formed. The process repeats itself and one cell is added to another. It cannot be denied that the pictures reproduced have a marked external resemblance to natural algse, fungi, etc., and that it is 256 COLLOIDS IN BIOLOGY AND MEDICINE possible to imitate nucleus and cell division, growth and all sorts of phenomena, and that even their internal structure is in many re- spects suggestive of a cellular structure. Doubtless structures do occur in nature which develop in the same way as these artificial osmotic products. In fruit wines after fermentation, H. Miller- Fig. 42. Microphotographs of osmotic structures, showing a cellular structure. Magnified X 60. (From St. Leduc.) Thürgaü* found vesicles which were filled with bacteria (Fig. 43) on the yeast sediment. These "bacteria vesicles" were developed because the colloid substances eliminated by the bacteria, in con- junction with the fruit wine, which contains tannin, form a semi- permeable membrane, a vesicle, that grows and sends out tubules. Is there really an analogy in the development of these structures to the development of natural organisms? ^ The fact that there is not the slightest chemical resemblance to organisms may be completely disregarded since St. Leduc and all who share his views speak only of the similarity of the physical force at work in both. 1 W. Roux has treated the entire question in a very instructive essay on the "Angebliche Kunstliche Erzeugung von Lebewesen" in the "Umschau" (Frank- fort a. M.), 1906, Nr. 8. GROWTH, METAMOliPHOSIS AND DEVELOPMENT 257 It is a detriment to the scientific treatment of tlie entire question that the outward resemblance (form and color) is so strikingly hke the natural structures. Involuntarily, we are reminded of wax figures which move their arais and legs. Though the internal struc- ture has some slight resemblance to some natural organisms, the analogy completely fails if a\ e consider such de- tails as cell division and cell multiplication. The figures of St. Leduc absorb no nourishment, other than water; their increased weight, as far as solid substances go, consists only in pellicle formation, and in spite of a shape suggestive of higher organisms, they have no differentiated internal structure, and the cell division has not the remotest resemblance to natural cell division, but occurs intermittently by bursting, etc. We might, if it did not seem fruitless, multiply the Yig 43 Vesicles of dissimilarities indefinitely and we might show Bacterium mani- that metabolism and the development of germ topoem from a cells is out of the question in such formations. P^^ culture in ■vrr i X 1 1 • 4.1, £ u sterile pear juice. We must not overlook, m the presence ot ail r^, • i i. j . ' 1 1 • 1 r. 1 he vesicle has de- these dissimilarities, that the physical forces veloped a long which produce these inorganic formations are transparent tube, the same as those ivhich produce the growth and Magnification configuration of organized material: membranes, 200:1. (After H. ,. j--ff • MüUer-Thm-gau.) osmotic pressure, dijjusion. ^ ' In one point, at least, Leduc's analogies fail completely: except- ing the membranes, they are devoid of colloid material. We have already seen that swelling frequently replaces osmotic pressure. If we could imagine the crystalloid material of St. Leduc replaced by colloids capable of swelling, we would have the essen- tial physical and chemical conditions for growth and structure of organisms. A serious study of these problems, one which extends beyond ex- ternal resemblances, is still in its earliest beginnings; see R. E. Liese- gang, Nachahmung von Lehensvorgängen. Formations resembhng the creations of St. Leduc have been independent^ noticed by others. B. D. Uhlenhuth* produced beautiful growths b}^ putting iron objects into antiformin. Antiformin is a mixture of sodium hypochlorite with sodium hydrate. The formations consist of iron oxid, and their development is easily understood from the ex- planations that have been already given. Since a small amount of a water-soluble iron salt must be formed first by the action of the hypochlorite of soda on the iron, the growth is slower, the figures 258 COLLOIDS IN BIOLOGY AND MEDICINE are more beautiful and possibly more natural. In two weeks the structures attain a height of from 5 to 10 cm. The growths described above offer analogies to organisms which are completely surrounded by water only, yet there are those which offer a resemblance to the growth of terrestrial plants. We might mention the ''blossoms" of many crystalloid substances, especially the ammonia salts. H. Wislicenus*^ has thoroughly studied the growth and structure of fibrous alumina. If granules of aluminium which have been activated by contact with traces of mercury subli- mate, for example, are permitted to he in a moist place, very soon Fig. 44. Fibers of fibrous alumina, magnified X 40. (From H. Wislicenus.) fibrous structures grow from the metal, which in a few hours may reach more than 1 cm. in length. Under the influence of the mercury as a catalyzer, aluminium oxid is formed from the aluminium accord- ing to the formula : Al + 3 H2O = Al (0H)3 + 3 H. In this instance it is not the osmotic pressure of entering water which bursts the films, thus bringing a fresh metal surface to the reaction, but the pressure of the hydrogen gas. Though in this case, as well, there are many gaps in the resemblance to natural organisms, nevertheless the fibers formed show certain resemblances to real fibers (see Figs. 44, 45). GROWTH, METAMORPHOSIS AND DEVELOPMENT 259 In the first place they resemljle most organic fibers by being doubly refractive (L. Jost*) even though this doul^le refraction is caused by different conditions than in organized structures. In contradistinction to natural fibers, the substance is isotropic; it is only its lamellated structure which produces that type of double re- fraction (H. Ambronn*) which is found in the siliceous envelopes of Pig. 45. Piece of doublj'- diffracting fibrous alumina. Magnified X 440. (After H. Wislicenus.) diatoms (e.g., pleurosigma and amphipleura) and probably also in tabasheer, the colloidal silicic acid which occurs in the internodes of some species of bamboo. The Genesis of Structures. What phase is for the physical chemist, cells and tissues are for the biologist. Like phases, cells and tissues are "portions of a structure separated from one another by physical interfaces" (Wil- helm Ostwald's definition of phase). The interface maj^ consist of an invisible transition layer (see p. 280). The interface is most evident when it consists of a visible membrane. Such a membrane whether visible or invisible is always a structure poor in water. At 260 COLLOIDS IN BIOLOGY AND MEDICINE the interface against air it may be created by desiccation. Within the organism, we must assmne that membranes develop similar to chemical precipitation-membranes. If we add silver nitrate to a solution of common salt, a precipitate of silver chlorid is formed. If we permit common salt and silver nitrate to diffuse together in a jelly, at the point of contact a membrane of AgCl develops. We con- sidered the results of this more thoroughly in the introduction to this chapter. It is now, therefore, merely necessary to recall that not only may crystalloids form such membranes in a jelly, but that with albuminous material H. Bechhold*^ produced such membranes in jel- lies (phosphoric acid, goat serum and goat-rabbit serum). Theoreti- cally, therefore, the development of membranes offers no difficulties. The development of a precipitate in a jelly gives a certain direction to the further evolution of the process. The sense in which this is intended will be elucidated by some examples; by membrane forma- tion, to begin with. Substances which form no membrane by precipi- tation, diffuse together unhindered, and in time become completely mixed. If a semipermeable membrane has formed, it behaves like a solid wall that arrests any further mixture. If two solutions of equal osmotic pressure diffuse together in a jelly till they form a permeable membrane, no matter how thin, e.g., sodium chlorid and silver nitrate, diffusion ceases as soon as the membrane has a very slight thickness. If, however, the osmotic pressure on one side is greater, the mem- brane continues to grow until the osmotic pressure is equal on both sides (N. Pringsheim,* H. Bechhold and J. Ziegler*^). The phenomena are exceedingly interesting when precipitates de- velop simultaneously in several places. These phenomena have been studied by R. Liesegang.*^ If we place on a plate which is covered with sodium chlorid jelly, a drop of silver nitrate, there forms a disc-shaped precipitate of silver chlorid, whose circumference in- creases equally in all directions (a circle) according as the silver nitrate diffuses into the sodium chlorid jelly. If, however, two drops are placed on the sodium chlorid jelly several centimeters apart, there develops a picture like Fig. 46; the two silver chlorid precipitates grow towards each other, that is, an ''apparent chemical attraction" is observed. The reason for this is as follows: im- mediately upon applying the silver nitrate, the jelly loses sodium chlorid because of the precipitation of AgCl, and this causes a movement in the entire mass of sodium chlorid: the spot where the precipitate forms is deprived of chlorin ions, which then diffuse in afresh from the periphery. If two neighboring drops of silver nitrate have been placed on the sodium chlorid jelly, there forms be- tween them a region poor in chlorin, which thus permits a more rapid Fig. 51. Laminated urinary calculus Fig. 50. Liesegang 's rings. (From (urates). (Drawing by H. Schadde; R. Liesegang.) horn von Frisch and E. Zuckerkandl.) Fig. 53. Oil-droplets super-saturated with Cho- lesterin. Crystalline separation of Cholesterin may be recognized in several droplets. Fig. 52. Primitive gall-stone pattern. (Myelin clump.) Magnified X 62. Fig. 54. Laminated calcium bilirubin stone. PLATE II. GROWTH, METAMORPHOSIS AND DEVELOPMENT 261 advance of the silver nitrate. What is shown here in the case of the silver chlorid holds for every other precipitate, and for every osmotic disturbance, provided only that dif- fusible substances are present in a jelly. If such disturbances (precipi- tates, membranes) occur simultane- ously in various places, it is possible « ., , T J. 1 r: J. Fig. 46. Apparent chemical attrac- for the most complicated fagures to ^. ,-r, t- ^ ^ ° tion. (R. Liesegang.) form. In addition, we must consider the changes which occur in the ordi- nary course of swelling and shrinking of the colloidal material. As far as I know, it has hitherto not been possible to explain such a phenomenon in vivo. I wish to refer to only one analogy: C. U. Ariens-Kappers* described as neurobiotaxis a phenomenon of nerve fibers: if two nerve cells, a certain distance apart, are injured simultaneously or in close succession, the growth of the chief dendron of both injured ganglion cells occurs in the direction of the other stimulated or injured cell. We thus have a growth towards each other, analogous to the apparent chemical attraction just described. [C. A. Elsberg concludes from his experiments that hyperneuroti- zation of a normal muscle is impossible. A normal muscle cannot be made to take on additional nerve supply. The implanted nerve can- not make neuro-motor connections. If the muscle is permanently sep- arated from its original nerve, the implanted nerve will then establish such connections. Science N. S., Vol. XLV, p. 319 et. seq. Tr.] Naturally, cells mutually modify each other's shape. A structure which would develop spherically if uninfluenced, under the pressure of neighboring cells acquires a reticulated, fibrous or pavement shape. Layered Structures. We saw that if two solutions which form a precipitate meet in a jelly, a precipitation membrane develops at the point of contact. Provided this is sufficiently permeable and one solution has a higher osmotic pressure, the membrane continues to grow uninterruptedly, becoming constantly thicker until the osmotic pressure is the same on both sides. In 1898, R. E. Liesegang*^ pubhshed an observation which does not accord with the continuous growth mentioned above. If, for instance, ammonium bichromate is dissolved in melted gelatin, which is then solidified in shallow dishes, and upon it a drop of silver nitrate is placed, then there does not develop upon diffusion a constantly thicker precipitation membrane of silver Chromate but con- centric rings called Liesegang's rings (see Plate II, Fig. 50). The 262 COLLOIDS IN BIOLOGY AND MEDICINE experiment may be performed by solidifying the ammonium bichro- mate gelatin in a test tube and layering some silver nitrate over it. We thus get instead of rings true precipitation membranes, which are separated from one another by layers containing no silver Chro- mate (see Fig. 47). Subsequently, Wilhelm Ostwald, *2 J. Haus- mann,* H. W. MoESE and G. W. Pierce,* H. Bechhold*^ and E. Hatschek studied the development of these rings. It may be assumed as a result of these investigations that the formation of such layers is the result of a very complicated combination of events whose further elucidation at this point would carry us too far afield. It should be definitely stated that the de- velopment of rhythmic structures is in no way dependent on the interdiffusion of two solutions. Similar structures may be produced in jellies also by crystallization (e.g., tri-sodium phosphate) or by freezing water. The ring formation occurs especially when ammonium Chromate and silver nitrate come in contact; at times there may be produced as many as twenty or more parallel membranes, which, according to the concentration of the solution, may be separated from a fraction of a millimeter up to 1/2 centimeter. They have also been produced by numerous other precipitation reactions. H. Bechhold prepared similar stratified mem- branes with organic material. This occurs readily if serum mixed with gelatin is permitted to solidify in a test tube and metaphosphoric acid is layered over it. The number and beauty of the mem- branes depend very much on the relative con- centrations of the solutions employed. Most advantageous is a mixture of 2.5 per cent serum and of 5 per cent gelatin, upon which is placed 2 per cent metaphosphoric acid; this gives as many as five concentric rings. The author obtained two parallel membranes by the diffusion of goat serum into gelatin, containing goat-rabbit serum. It is evident in the formation of this kind of layered membranes that the phenomenon noted on page 85 et seq. plays an important role: colloids only precipitate in definite mixture relations, and solution oc- curs in the presence of an excess of either one. This phenomenon can be followed visually in the above-described membrane formation. jaty-^- Fig. 47. Stratifica- tions in a test tube. (F. Stoffel.) GROWTH, METAMORPHOSIS AND DEVELOPMENT 263 It is easy to imagine that layered membranes develop by the re- moval of a substance which holds another substance in solution. I have experimented with this end in view by dissolving globulin in gelatin containing sodium chlorid and layering water over it; when the sodium chlorid diffused away, then the globulin precipitated out. As a matter of fact no layered structures developed in the gelatin but only turbidities uniformly distributed. This does not by any means mean that, with a different arrangement of the ex- periment, regular layered membranes might not be obtained. In organisms we frequently encounter stratified structures which, in most cases, occur as the result of rhythmic deposition. The an- nual rings of trees, the various layers in the otoliths of young and of old fishes cannot be explained in any other way than that periods of Fig. 48. Starch granules. (Kilnitz-Gerloff.) rest follow periods of strong accretion. In contrast to these "external rhythms" which obviously are induced by changing conditions affect- ing an organ, there are also layered structures with "internal rhythms " which suggest Liesegang's rings As such, we must regard starch grains (Fig. 48), the sihcious, sponginous and calcareous structure of sponges, and the perforated calcareous shells of the foraminifera and many fish scales. R. Liesegang*^ mentions in addition the con- centric lamellae about the Haversian canals in the bones of verte- brates, the rods of the retina and the spiral cross strise of muscle fibers. W. Gebhardt compares the rhythmic markings on butterflies' wings to Liesegang structures. The coarse layers of the otoliths of fishes, the annual rings of trees, the concentric structure of pearls are penetrated by still finer layers, whose formation R. Liesegang thinks is analogous to the formation of the rings he described. 264 COLLOIDS IN BIOLOGY AND MEDICINE E. Küster has called attention to numerous rythmic phenomena among plants in which the influence of external forces cannot be repognized and which consequently he attributes to Liesegang's law. Among others we may mention the thickening of membranes in vessels and trachese, the bands in striated portions of plants (herbs, pinus Thunbergii Pari.), and the rhythmic changes in many blos- soms. Doubtless the number of instances where we may entertain the idea of internal rhythms may be increased at will and it will be the task of future investigators to determine the essential causes. A valuable contribution in this connection was made by M. Munk who studied the formation of fairy rings. When moulds are grown on bread, nutrient agar, etc., we frequently observe growth in con- centric rings which suggest Liesegang's layers and are popularly called fairy rings. M. Munk demonstrated that the accumulation of metabolic products interfered with growth, causing a zone where there was but little mould. In the case of some strong acid producing moulds the distance between the individual rings may be regulated by the addition of alkali to the nutrient medium. If litmus agar is used the blue and red rings make visible the cause of the ring formation. It is interesting that layered structures in the ends of the periph- eral nerves, which were looked upon as real, have been proved to be artefacts. Golgi stained nerves by saturating them with potas- sium bichromate and then treating them with silver nitrate. He obtained stratified structures whose appearance, as H. Rabl showed, changed with the concentration of the solutions, and they could be nothing other than Liesegang's rings. Biological Growth. The fertilization of the egg is evidently the cause of the powerful swelling processes, which are possibly induced by the formation of acids. According to Jacques Loeb,*^ oxidation processes accompany the development of the egg (whether fertilized or parthogenetic) ; without oxygen no development of the ovum occurs. The increase in the volume of the ovum of Echinoderma, until it reaches the pluteus stage, is entirely conditioned by the absorption of water (C. Herbst*). Before the larvae reach the pluteus stage they can- not a.ssimilate any organic nourishment. Davenport* in the case of frog embryos has shown that their dried weight remains the same or diminishes till the moment when they commence to eat. Their water content on the other hand was enormously increased. This absorption of water is not due to an increase in the osmotic pressure^ GROWTH, METAMORPHOSIS AND DEVELOPMENT 265 since the fertilized and the nonfertihzed frog's egg shows, according to L. Backmann and J. Rünnstrom,* only 1/10 of the osmotic pressure of the egg in the ovary, or of the adult frog. In the course of development, the osmotic pressure increases so that in the tadpole of 25 or 30 days, the osmotic pressure is almost the same as in the metamorphosed animal. L. Backmann and J. Rünnstrom agree that the decrease in osmotic pressure is due to the fertilization, which results in a gel formation by means of which crystalloids are adsorbed. After a certain time, which varies for different animals, but not as yet definitely established for individual ones, a shrinking begins again, as may be seen from the following data, taken in part from the tables of H. Gerhartz*^: Man. Water, per cent 94.0 90.0 86.0 83.3 71.7 67.6 66.0 Dry substances, per cent. 3d fetal month 6th fetal month (Rubner) 7th fetal month (Rubner) 8th fetal month (Rubner) Newborn (Camerer, Jr. ) . . Adult (Moleschott) Adult (Bouchard) Fetus (I inch long) (A. v. Bezold) . Newborn (A. v. Bezold) 8 days old (A. v. Bezold) Full grown (A. v. Bezold) 87.2 82.8 76.8 73.3 6.0 9.7 14.0 16.7 28.3 32.4 34.0 Dog. 6 days old (Gerhartz) 80.3 77.0 19.7 15 days old (Gerhartz) 23.0 Sheep. 6 months old (Lawes and Gilbert) 47.8 43.4 52.2 15 months old 56.6 Mouse. 12.8 17.2 23.2 28.7 Chicken Embryo (without yolk). 7th day (L. v. Liebermann) , 14th day (L. v. Liebermann) . 21st day (L. v. Liebermann) 7.2 12.7 19.65 What constituents are especially deprived of water cannot be properly determined from the limited material at hand, yet, accord- 266 COLLOIDS IN BIOLOGY AND MEDICINE ing to H. Gerhartz, there seems to be a very great shrinking even of the albumin. For man he calculates the proportion of albumin to water to be: Newborn 1 albumia, 5.6 water Adult 1 albumin, 4.3 water J. A. KuBowiTSCH has shown that the water content of a mam- malian embryo's muscle sinks from 99.4 per cent to 8 per cent at the termination of fetal life and finally to 75-80 per cent in adults. According to L. B. Mendel and Leavenworth, pig's hver has a quite constant water content of about 80 per cent during fetal life but diminishes to 67.3 per cent in the adult. From this we understand that in the earliest stages, growth only results by means of the water taken up through swelling, though a time comes when growth is induced by the entrance of solid sub- stances, by assimilation. This assimilated substance meanwhile binds less water; with further growth there is associated a relative shrinking which after reaching its maximum (growth) passes with further age into an absolute shrinking. According to Mühlmann, aging of different organs does not proceed equally. The weight of human intestines increases up to the fiftieth year, but the heart and lungs never cease gaining weight; the brain, on the other hand, has achieved its maximum weight at about the end of the second decade and from then on it gradually declines. The brain also shows definite microscopic aging phenom- ena, even in the earliest years. Lipoid pigment granules appear in the nerve cells which continually increase and at an advanced age fill the entire cell According to Marinesco it is much easier to destroy with solvents suspensions of ganglion cells of a newborn puppy than an old dog. As the result of his studies of pigment granules in nerve cells he also arrives at the conclusion that aging is due to the coagulation of physiological elements, a diminution of surface tension, such as we know occurs in the aging of colloids (see p. 73). H. Schade determined that the subcutaneous connective tissue dis- solved much more rapidly in NaOH when it was derived from a month old child than when it was taken from a thirty-two year old woman. F. Tangl* is of the opinion that the shrinking of the animal organism during embryonal development is a duplication of the same phylogenetic process and shows by numerous tables that the lower invertebrates, even those which do not live in water, are usually more rich in water than the higher vertebrates. By what chemism increase of water and substance are condi- GROWTH, METAMORPHOSIS AND DEVELOPMENT 2()7 tioned, how cell dividon results, what are the relations between nucleus and cell protoplasm, are questions which are not yet ripe for colloid research. To be sure we know that such swelling and shrinking processes occur, not only for the entire cell, but also in the nucleus. Before each cell division the nucleus swells up very much, and after division shrinks again, decidedly. BoROWiKOW* made interesting observations on plant growth. All who are familiar with plants know that even in summer, periods of apparent rest alternate with periods of active growth (sprouting). The latter phase is associated with considerable entrance of water. It is impossible to explain this inhibition of water by osmotic forces since the increased rate of growth was usually associated with a diminution in the concentration of the cell juices instead of the reverse, and a growing plant absorbed unequal quantities of water from solutions osmotically identical. There was, on the contrary, some evidence that sweUing processes were active. Martin H. Fischer had already called attention to the fact that the tips of buds are always acid in reaction. It was quite natural, therefore, to test the influence of acids, bases and salts on the sprouting of plants and to compare it with the swelhng of colloids. For this purpose Boro- wiKOW placed six-day old sunflower seedlings (HeHanthus ammus) in sieves and dipped them in various solutions, using distilled water as a control. Dilute acids (1/100 normal) accelerated growth while salts simul- taneously present acted against the acids. Acids and salts were active in a series which was analogous to that for the swelhng and shrinking of dead colloids. That bases caused no acceleration of sprouting seems to mihtate against the original assumption. Borowikow explains this by the fact that the cell juice in the growth zone is essentially acid and con- stantly forms carbonic acid; the bases neutralize the acid, forming neutral unhydrated albumin and in higher concentrations damage the plants. In this way he explains the stimulatmg action of dilute solutions of organic bases (0.001 n urea nitrate, 0.0015 n caffein sulphate, 0.0025 n phenylene diamine chlorid) which, according to Borowikow, act like their respective acids since they are hydrolyzed in solution. Borowikow expects especially to bring growth into relationship with turgor (tissue distention). Unnoticed, great turgor may be diminished by the growth process. According to Borowikow growth is ionization of the plasma protein by H ions in the gro^vth zone, causing the protein to pass from the gel to the sol condition. 268 COLLOIDS IN BIOLOGY AND MEDICINE Ossification Processes. One of the most interesting of colloid-chemical problems is bone formation. We shall see on page 302 that from an aqueous solution containing blood salts, calcium carbonate and calcium phosphate precipitate. The precipitation is hindered by the presence of the blood colloids, though two-thirds of Ca salts, at least in the serum of higher animals, occur in the crystalloid state. This interference must stop during the formation of bone. To account for this there are several theoretical possibilities : it may be assumed that changes in the serum colloids are brought about at or from the bone cells, which remove their protective action and results in the precipitation of the calcium salts. This agrees with the views of Wo. Pauli and Samec,* which we shall consider more closely. It was shown in their researches that the increase in the solubility of calcium carbonate by serum albumin was 475 per cent, and of calcium phosphate 90 per cent. We would consequently expect to find a very much more extensive pre- cipitation of calcium phosphate than of calcium carbonate when the protective action was removed. But in the case of bones, the pro- portions are just the reverse. The bone ash of man contains about 850 parts Ca3(P04)2 and 90 parts CaCOs per 1000. But in the case of a cleavage product of albumin. Wo. Pauli and Samec found that the solvent action upon calcium salts was the reverse. Witte' s peptone, consisting almost entirely of albumoses, holds in solution only the calcium carbonate, whereas the calcium phosphate exhibits a diminution in solubility. Based on these results, ossification might occur in the following way : In the bone or cartilage cells, there occurs a concentration of colloids in which a large quantity of calcium salts are piled up. When these tissue col- loids are broken down, a precipitation occurs, the precipitate consist- ing chiefly of calcium phosphate with smaller amounts of calcium carbonate. This corresponds with the histological evidence, by means of which a tissue destruction may be seen to accompany ossification. A further possibility, which does not in the least contradict the above explanation but possibly coincides with it, is that phosphates are set free and come into contact with the carbonates always present when the tissues, especially the cell nuclei, break down. In accordance with well-known physico-chemical laws, an increase in the concentration of an ion (in this case the phosphate ion) results in an increase in the calcium phosphate molecules, and this changed albumin must, accordingly, favor the precipitation of calcium phosphate. Finally we may think of a kind of specific adsorption by certain cell groups. In fact, M. Pfaundler noticed a selective adsorption of GROWTH, METAMORPHOSIS AND DEVELOPMENT 269 calcium when he placed pieces of cartilage in chloric! of üme solution. This suggests the method by which hme is deposited in damaged tissues (vessel walls or tubercles). We must also consider the simultaneous precipitation of positively and negatively charged albumin with the breaking down of calcium salts as we shall describe later when we discuss concrement formation at greater length (see p. 271 et seq.). Finally, we may consider some kind of specific adsorption by definite cell groups. We might also consider the mutual precipita- tion of positively and negatively charged albumin which carry salts down with them in the same way as is more fully described in the case of concrement formation (p. 271 et seq.). What has been said here of bones also holds, of course, equally well in principle for the shells of mohuscs and snails, for the armor of crustaceans, as well as for other ossification phenomena. Morphol- ogists distinguish primarily between calcification and ossification (see Gebhardt) . In lower animals (shells of snails and mussels, carapace of crabs, spicules of sponges, etc.), the Hme salt occurs chiefly in microcrystalline form, as fine granules in the calcifying tissues. In contrast with this calcification, lime forms an optically completely homogeneous deposit in bone and never occurs as a formed or crystalhne precipitate. Possibly this essential difference depends on the fact that special cells, osteophytes, take part in bone formation. It is still impossible for colloid research as yet;, to offer even an hypothesis in explanation of this difference. R. Liesegang* (his correctness in doing so, I shall not discuss) criticises an explanation of ossification which involves the presence of special cells, the osteoblasts. He calls attention to the fact that deposits of lime occur in places where there are no osteoblasts, as in the arterial wall, in arteriosclerosis, or in brain cells. He evidently concludes that under some circumstances, even without a special storing- up, it is possible to have a precipitation from blood serum supersaturated with calcium salts, in which action the formation of centers or nuclei possibly take part (similar to the theory of H. Bechhold and Ziegler*^ for the deposition of urates). The very marked density and the poverty of the bony framework in organic substances is deserving of special consideration. For this, the investigations of R. Liesegang*^ offer valuable experimental support. He showed that when calcium phosphate membranes were allowed to form in gelatin jellies (by the diffusion of disodium phosphate and calcium chlorid towards each other), that they were almost free from gelatin; to a certain extent the organic supporting substance had been forced away. 270 COLLOIDS IN BIOLOGY AND MEDICINE This investigator has simulated the formation and growth of the long bones. He filled test tubes half full of gelatin which was made alkaline, with tricalcium phosphate, for instance; this layer repre- sented the periphery of the bone. After solidification there was placed on top a thin coating of gelatin containing a suspension of tri- calcium phosphate; this was to represent the bone. Upon this was poured some acid solution, for instance, lactic acid; representing the center of the bone. The acid diffuses through the tricalcium phos- phate layer, dissolves the calcium, reaching the lower periphery, where the calcium is precipitated again in layers as a phosphate. If a suitable calcium salt is added with the lactic acid, the layer becomes stronger and less porous; in a successful experiment, it shows, inside, the characteristic worm-eaten appearance of the long bones, and out- side, a smooth firm and sharply defined structure. As natural sources of acids, R. Liesegang mentions the accumulation of CO2, lactic acid and glycerophosphoric acid derived from lecithin. [Barille found that tricalcium phosphate was dissolved by water containing CO2 under pressure, forming an unstable compound tribasic calcium carbon phos- phate, CasPsOs + 4 H2CO3 = H2O + PaOsCaa : 2 CO (CO3H) Ca.] In discussing calcification and ossification in his Harvey Lecture, 1910-11, H. Gideon Wells concludes that "there seems to be no essential differences between the processes involved in normal ossifi- cation and in most instances of pathological calcification. Any area of calcification may be changed to true bone in the course of time," and that "' calcium deposition seems to depend rather on physico- chemical processes than on chemical reactions." Jerome Alexander suggests the importance of the removal of or alteration of protective substances resulting in the deposition of calcium salts. Cf. also Hofmeister' s observation on the difference between solubility of calcium phosphate dissolved in serum and the dissolving of calcium phosphate by serum. Tr.] Diseases of the Bone. Of the noninfectious bone diseases, rickets and osteomalacia attract our special attention. Rickets is characterized by lime-poor, so-called osteoid tissue, in- stead of the solid calcareous structure. In this way a pliable mass takes the place of the rigid framework. This lack of lime might readily be attributed to a lack of lime in the food, but it has been shown that this is certainly not the cause, since such a lack of lime can be produced only by artificial preparation of the food. In my opinion Pauli's theory of bone formation offers a good explanation for rickets. He supposes, as indicated on page 268, a prehminary GROWTH, METAMORPHOSIS AND DEVELOPMENT 271 tissue degeneration; in rickets we find quite the reverse, namely, an over-production of tlie osteoplastic tissue, so that we lack the conditions necessary to the precipitation of calcium phosphate and carbonate, or, in other words, bone formation. Osteomalacia, bone consumption, is in certain respects the reverse of rickets. If in rickets we find a deficient precipitation of insol- uble lime salts, in osteomalacia we have the eating away of existing bone. Osteomalacia occurs most frequently during pregnancy, dur- ing which even under normal conditions the teeth may suffer. Osteoporosis, the bone consumption of the aged, which is especially noticeable in the skull, belongs to this group. Here, too, we must reject the theory of the deficient introduction of lime salts, as it is contradicted by all metabolism researches. We are much more in- clined to accept a dissolving away of calcium phosphate and carbon- ate, especially by acids. Since the oxidizing processes are deficient and the circulation functionates less perfectly, an accumulation of acids is not surprising. Magnus Levy has raised the objection to this ''acid theory" that the proportion of the calcium phosphate to calcium carbonate is the same in osteomalacic as in normal bones. He placed normal bones in lactic acid and found that much more carbonate than phosphate is dissolved away, and from this he con- cluded that the ''acid theory" was useless. This objection can- not be allowed. If acid diffuses from any direction into a mixture of calcium carbonate and phosphate imbedded in a jelly, the acid advances only to the extent that it has previously dissolved away all the carbonate and phosphate; this was shown experimentally by R. LiESEGANG*^ (assuming, of course, that an acid stronger than phosphoric acid is employed). As may be readily seen, the result of the experiment depends entirely upon the conditions; at any rate the contribution of Magnus Levy cannot count against the "acid theory." Concrements. In various pathological processes we find in the body cavities of animals and men, structures varying in size from that of a grain of sand to that of a fist, and which have developed without the help of cells. Such precipitates are called concrements. We find them as renal gravel, urinary calculi, gallstones, brain sand (in the lymph spaces of the brain), rice bodies in the exudate of diseased joints; as the pearl of the pearl oyster; and similar formations which are found at times in cocoanuts, in view of their structure, can be considered nothing else. The common characteristic of all concrements is that in addition to the special characteristic ingredients (urates and Cholesterin) they 272 COLLOIDS IN BIOLOGY AND MEDICINE always contain albuminous elements and they usually show a scaly and radial structure. These formations have been studied with especial care by H. Schade* as well as by L. Lichtwitz. *i In the following pages, we shall consider the origin of urinary and biliary calculi from the standpoint of these studies. Urinary Calculi. H. Schade mixed ox plasma, which had been made uncoagulable by the addition of potassium oxalate, with an emulsion of calcium phosphate and calcium carbonate. When he coagulated the mass by the addition of CaCl2, a hard cake formed, which, when preserved in salt solution at 40°, shrank, and after eight weeks had approximately the hardness of a fresh urinary calculus. Fibrin was absolutely necessary, yet from 0.07 to 0.1 per cent in plasma diluted ten times sufficed to produce a coagulation, and the phenomenon was the same if neutral urine was used for dilution instead of physiological salt solution. By changing the composition of the sediment (mineral ingredients) it is not difficult to produce stratified structures resembling urinary calculi. This similarity is not a mere superficial one. Renal calculi have been repeatedly found which were still soft and plastic like the initial stages of these artificial stones. Whether the organic layered framework of natural urinary calculi (see Plate II, Fig. 51) consists of fibrin is still an open question, though there is much in favor of this view. According to H. Schade the formation of urinary calculi is somewhat as follows: coagulum and urates sediment simultaneously or in close succession, shrink and harden. By the repetition of such processes layers form, the stone grows, and after a while becomes stony hard, because the crystalloid ingredients grow into large crystal aggregates and take on a radial structure. The lesson for therapeutists is, that not only must the formation of crystalloid sediments be prevented, but also the passage of fibrin or similar colloids into the urine. Alone, urinary sediments form a crumbling mass. Calculus formation is possible only by means of colloidal "mortar." Gallstones (biliary calculi): Gallstones differ very widely in their chemical composition. We recognize those which merely consist of Cholesterin and others which contain only calcium bilirubin; between these extreme forms occur all sorts, consisting of mixtures of these two chief constituents with albuminous material. Without going into the individual reasons, it may be said that, according to H. Schade, Cholesterin appears to be dissolved in the bile as a hydrophile, and calcium bihrubin as a hydrophobe colloid. It must also be noted that, besides the chelates, the bile contains salts of the fatty acids, lecithin and mucinous substances which GROWTH, METAMORPHOSIS AND DEVELOPMENT 273 must partly be regarded as solvents for the gallstone material. Cholesterin precipitates in individual crystals from a supersaturated aqueous solution of etiolates; however a few drops of an oil suf- fice to cause precipitation in an amorphous clump very similar to the ''myelin clump" which Naunyn described as the (uranlage) '' precursor of gallstones" (Plate II, Fig. 52). After a few days, radiating crystallization starts from the center (Plate II, Fig. 53), oil droplets are released and may again give rise to oil-cholesterin precipitates. In this way may be prepared, artificially, hard or more or less plastic Cholesterin stones, such as rarely occur, however, in the gall bladder. The presence of fats or fatty acids is, therefore, a requisite for their formation. For the precipitation of Cholesterin it is only necessary that the substances which hold it in solution, the cholates and salts of the fatty acids, be destroyed. A priori, a supersaturation with Cholesterin is brought about in all processes that interfere with the normal alkaline reaction of the bile, especially infection with B. coli, B. typhosus, B. pyocyaneus and B. proteus. I am inclined to accept the view of L. Lichtwitz*i that the acid for- mation of these bacteria is chiefly responsible for the breaking down of cholates and soaps, since staphylococcus aureus, which does not form acid, causes no separation of Cholesterin. The precipitation of Cholesterin can also be caused by sterile autolysis as well as by rendering less favorable the conditions requisite to solution, since cholates are absorbed by the walls of the gall bladder if the bile stagnates there (congestion). Bilirubin forms amorphous precipitates with lime salts which normally do not sediment out in the bile. In the presence of al- bumin and fibrin, under conditions as yet not accurately studied, calcium bilirubin may precipitate and include the albuminous in- gredients, giving rise to clumps which in their cheesy stt-uctures are very like natural calcium bilirubin stones (Plate II, Fig. 54). In my opinion the neutral or faintly alkafine reaction of the bile is essential for the development of a calcium bilirubin stone as op- posed to the Cholesterin stones, which require acidity. H. Schade* considers that catarrhs, inflammatory and strongly exudative proc- esses in which much lime enters the bile are responsible for the formation of calcium bilirubin stones; this view explains the occur- rence in them of albuminous ingredients. In some of the so-called mixed forms there may be an alternation of processes which condition the separation of Cholesterin and of calcium bilirubin. An answer to the question whether a simultaneous precipitation of Cholesterin and calcium bilirubin may occur would be very interesting. 274 COLLOIDS IN BIOLOGY AND MEDICINE H. Schade rightly sees in every section through a stone, a ''valu- able document written by nature on the development and course of gallstone disease. " Gout. Gouty deposits, tophi, which form in the joints by preference, offer a certain analogy to ossification processes. In uric acid arthri- tis as well, the blood is from time to time supersaturated with a difficultly soluble salt, monosodium urate, which precipitates under favorable conditions. Gout is regarded as a disease of nuclein metaboHsm. We must forego a discussion of this part of the question at present, and limit ourselves to determining the influence exerted by the colloids of the organism upon the deposition of urates, the most characteristic phenomenon in the disease picture of uric acid arthritis. Two papers by H. Bechhold and J. Ziegler*'' as well as by F. Gudzent,* which appeared in 1912, have given a certain direction to these views. Until the appearance of these papers, no stress was laid on the question whether uric acid appeared as such or as urates in the organ- ism. The reason for this was that all analytical investigations deter- mine the uric acid, and since it was known that uric acid is much less soluble in water than its alkaline salt, it was taken for granted that the same condition held for the fluids of the body, especially for the serum. H. Bechhold and J. Ziegler showed, however, that no free uric acid but only urates, chiefly sodium urate, exist in the body, and they showed, moreover, that sodium urate was much less solu- ble in serum than is uric acid. The subsequent publication of F. Gudzent confirmed and explained this experimental result by the electrolytic dissociation of the electrolytes in question. F. Gudzent started with the idea that albuminous substances play no part in these processes, but H. Bechhold and J. Ziegler have shown that this assumption is erroneous. A few figures will explain what has been stated. Solubility (at 37° C). In water. In serum. 1 : 15500 1 : 665 Uric acid J 1 : 1925 (a) 1 1 : 40,000 (6) Monosodium urate^ 1 In dissolving sodium urate in these two ways, we do not obtain the same equilibrium. Only 1 : 40,000 monosodium urate dissolves in serum (&), but if monosodium urate is permitted to form by dissolving uric acid in serum (a), then 1 : 1925 dissolves. [This is probably due to the protective ac- tion of the serum, as the result of which some of the sodium urate remains colloidally dispersed. Tr.] GROWTH, METAMORPHOSIS AND DEVELOPMENT 275 It follows from this in contradiction to previous views that the blood in gout is frequently supersaturated with monosodium urate. According to the analysis of G. Klemperer, Magnus-Levy and SalOxMon, the uric acid content of the blood in gout varies between 30 and 80 mg. per liter, whereas in normal blood at most only traces of uric acid can be demonstrated. When the content reaches 25 mg. of sodium urate per liter of blood serum, every further addi- tion must be associated with a deposition of sodiimi urate, provided urate nuclei are present. We thus see that the serum colloids are of great importance for the solution of sodium urate in the blood and in preventing its deposition in gouty processes. [Stanley R. Benedict in his Harvey Lecture, 1915-1916, p. 362, discusses the presence of two forms of uric acid in blood. He determined ten times the amount of uric acid originally obtained by the preliminary boiling of the protein free filtrate with hydrochloric acid. The prob- able destruction of a '' protective " substance is quite apparent. This aspect has an important bearing on uric acid determinations in nephritis and out. Tr.] The influence on these processes exerted by radium emanations which inhibit the deposition of sodium urate from supersaturated serum (H. Bechhold and J. Ziegler**) deserves the attention of students of colloids. CHAPTER XVI. THE CELL. It is said that the cells are the structural units of the body; this comparison, however, is valid only to a very limited extent. Build- ing stones do not vary much in form and structure, and especially in their individual uses; the cells of the body, however, are of such manifold appearance and have such numerous uses that it is very difficult to discover what is common to the various kinds. A cell may consist of protoplasm, the cell contents and a nucleus; there may even be no nucleus. In plants, there is usually a visible cell pellicle which supports the protoplasm; this occurs less frequently in animal cells. Protoplasm, nucleus, and pellicle are, however, the microscopically distinguisha- ble parts. Theory requires an invisible plasma pellicle which is con- sidered the surface of the protoplasm. There can be no doubt that a cell consists of many elements beneath the limits of visibility, which are responsible for definite functions. Independent microorganisms exist which are invisible because of their size (the ultramicroscopic causes of some diseases, e.g., smallpox, measles, etc.), which possess all the properties of an independent complicated organism (nutrition, propagation, etc.). We shall endeavor to get an idea of the molecular structure of a cell from an analogy of F. Hofmeister who took as his example a liver cell, which performs particularly numerous functions. It occu- pies approximately the space of a cube with edges 20 /x long = 8000 /J = 8'10~^ mm^. Assuming the following conditions F. Hofmeister arrives at the subsequent figures. A gram molecule of any chemical substance consists of 0.62 quadrillion (0.62« 10^*) molecules. From this 0.62- 10^* we may calculate the number of molecules present in a definite space if we know the weight and composition of a cell. F. Hofmeister assumes that for colloids, on the average, the molecular weight is that of hemoglobin (about 16,000), that for lipoids 800, and for crystalloids 100. Consequently, we must figure that a liver cell contains 76 per cent water 225,000 milliard molecules^ 16 per cent protein 53 " " 2^ per cent lipoids 166 " " 5| per cent crystalloids 2,900 " " 1 [1 milliard = 1000 millions. Tr.] 276 THE CELL 277 In order to grasp these figures F. Hofmeister shows how a struc- ture might be erected whose molecules are bricks, not to exceed in number 200,000 miUiards, of which 200 milliards colloid molecules with a portion of the salt molecules form the walls, roof, ceihngs, etc., whereas the water molecules with the remaining crystalloid molecules fill the rooms, halls and corridors. If such a structure had the enormous average height of 50 meters it would cover a ground space 7000 square kilometers or one-half the area of Alsace-Lorraine. It is evident that the compexity of the molecular structure of a cell baffles our powers of description. A cube with edges O.l^u which is much smaller than the Hmits of microscopic visibility contains 25 million molecules of water, 25 thousand molecules of colloidal, and 250 thousand molecules of crystalloidal substance, which under the same conditions would cor- respond to a building 100 meters front, 20 meters high and 20 meters deep. Protoplasm. Until recently there was little definite knowledge concerning the colloidal nature of protoplasm, that is, whether it was fluid or gela- tinized. It was known that after the fragmentation of yeast cells it was possible to press out a juice containing various enzymes, and that meat juice obtained in a similar manner contained albumin. In the case of yeast it may be inferred that the protoplasm contains sols, but in the case of muscle such an inference is met by the objec- tion that the albuminous substance may have arisen from the blood serum which bathes the muscle fibers. The facts that portions of cells form drops and that foreign fluids in protoplasm assume spherical shapes likewise point to the fluid nature of many protoplasms. Most of the numerous investigations concerning the physical nature of protoplasm are at present of mere historical interest, since the ultramicroscope has solved many of the main questions or has placed us in a position to do so in the future. One of the most important criteria for differentiating between a sol and a gel ^ is the presence of Brownian movement. If it is possible to observe an oscillatory movement in the granules of a cell, such granules must be in a fluid medium; if they are motionless the medium must be either a gel or very viscous. If we observe that the oscillating movement has ceased, it means that the fluid has gelatinized. Numerous ultramicroscopic observations of cells have been pub- lished. Plant cells have been studied most carefully by N. Gaid- UKOV.* He studied the pollen hairs of tradescantia, m3':xom5^cetes (shme fungi), the ceils of various algae (spirogyra, cladophora, oedogo- 1 [There is no sharp Une between sol and gel. The more viscous the medium the longer time the changes take, vid- metals for examples. Tr.] 278 COLLOIDS IN BIOLOGY AND MEDICINE nium, desmids, diatoms, oscillaria, etc.), yeasts, the common frog- bit, vallisneria, and several mosses. Gaidukov comes to the following conclusion as the result of these investigations: Protoplasm must consist of hydrosols because everywhere in living protoplasm he saw particles with Brownian movement. Frequently he observed that the particles combined or separated and that their number increased or diminished, phe- nomena which in my opinion represented metabolic changes. In some, usually in very well nourished cells, there were no movements, which may be attributed to the fact that the distances between the numerous particles were too small. A transition from sol to gel condition, i.e., the cessation of Brownian movement, was not observed in normal living cells. The colloids of plant protoplasm evidently consist of a reversible and an irreversible portion. If a cell is injured so that protoplasm escapes, a portion will expand in the water and ultimately be dissolved in it, whereas other portions combine (precipitate). This was observed in living and in dead protoplasm. The observation of 0. Nägeli is analogous: If we crush a root hair of hydrocharis in water under a cover glass, clumps of protoplasm pass through the rent. They are immediately surrounded by a membrane which is impermeable for dyes; i.e., they are coagulated on the surface. There is formed at the site of the wound an irreversible layer of hydrogel similar to the fibrin forma- tion of higher animals. The ultramiscoscopic observations quoted entirely confirm what was earlier observed when plant cells absorbed water. For instance, if we place myxomycetes in water, they swell; and in spite of the increase in surface, the outer hyaloplasm retains its thickness. Evidently the entering water causes a gelatinization of the granular plasma at the surface of contact, so that the hyalo- plasm layer spontaneously supplements itself. W. W. Lepeschkin regards protoplasm as a loose combination of proteins and lipoids which breaks down under lethal conditions (coagulation). If I understand him correctly he does not assume the existence of a plasma pellicle (see p. 245) but believes that all the properties ascribed by other investigators to this membrane as a limiting surface should be attributed to the protoplasm as a whole. It must be remembered that all these observations experience certain limitations depending on the kind and the part of the plant involved. As a result of his experiments with sunflower seedlings, BoRowiKow assumes that plasma exists in seeds and spores as a solid phase which changes to the gel condition in resting plants (evidently jellies are meant). In the growth period, the plasma as the result of hydration exists THE CELL 279 in the gel condition, the one in which we usually find it in cells according to Borowikow. With the occurrence of death -protoplasm gelatinizes, Brownian move- ment of the smaller particles ceases, and the structure of the gel appears in the ultramicroscope as a conglomeration of many re- flecting platelets. It makes a substantial difference whether the protoplasm slowly dies or is suddenly killed by a fixative (alcohol, formalin, etc.)- In the first instance there is a precipitation (floccu- lation), whereas, in the latter there is a stiffening; this difference may be readily recognized under the ultramicroscope. From this we may understand why a dead plant cell simply bursts in water, for the defects are no longer repaired from within. The cell contents have been already gelatinized. Chlor ohlasts (chlorophyl granules) may be assumed to possess colloidal properties similar to protoplasm ; only it seems the latter are more delicate (Ponomarew) . The living protoplasm of many animal cells, however, seems to exist as a gel. At least in monocellular organisms, blood cells, etc., A. Mayer and G. Schaeffer* could not discover any Brownian move- ment of certain granules. On account of the great differentiation of animal cells, more com- prehensive investigations must be awaited; thus it appears to me probable that red blood cells have viscous contents (see p. 305). The Nucleus. We know even less about the colloidal nature of the nucleus than of protoplasm. Ultramicroscopically, the nucleus appears to be a complex of hydrosols containing larger particles and to be quite poor in water. This corresponds well with the picture produced by staining. The colloids of cell protoplasm seem to be rather indifferent chemically; they are poorly stained by both acid and basic dyes. The nucleus, or more properly the chromatin substance, seems to possess pronounced acid properties, which are manifested by its intense staining with basic dyes (see A. Kossel*). The Cell Membrane and the Plasma Pellicle. The cell pellicle imparts its shape to the fluid protoplasm which otherwise would be spherical as the result of surface tension. The cell pellicle occurs in plants especially. In animals an interior skeleton or a spongy framework may determine shape. Theory requires an additional invisible plasma pellicle as bounding the 280 COLLOIDS IN BIOLOGY AND MEDICINE protoplasm. Formerly from an interpretation of the experimental facts as pure osmosis, the pellicle was considered semipermeable, that is, permeable for water but impermeable for everything else. This view is theoretically untenable since the cell requires numerous substances and there must also be an exit for excreta. As a matter of fact, the studies of recent years have shown that the plasma pellicle is by no means as impermeable for many substances as was assumed. The attempt has been made to decide its chemical and physical condition from the nature of the substances which pass through it. The statements on page 281 show that there is no uniform opinion as to whether the plasma pellicle consists of albuminous or lipoid substance or a mixture of both. In my opinion it differs in each instance depending on the contents of the cell. However, colloid research offers at least a foundation for a conception of the plasma pellicle. In animal cells, with few exceptions, we can discover no membrane, yet many of their properties indicate that they also possess some sort of a pellicle. Colloid chemistry gives us a basis for the explanation of such phenomena. The conditions are most simple when the cehs are surrounded by air. We know from page 33 that colloids concentrate and unite into a firm skin at an interface, fluid/air. The process is much more complicated in the case of cells in a fluid or semifluid medium. Let us recah the following experiment: If ether is shaken with water containing albumin or albumose, there will form a foam consisting of drops of ether surrounded by albumin or albumose films. Certain other colloids and fluids permit the formation of fluid foams, which have unmistakable similarities to agglomerations of cells. Although this analogy may at first sight seem entirely superficial, we must remember that the interface between two immiscible fluids and between a fluid containing solid or semisolid bodies (gel), possesses other properties than does the interior (see pp. 14-17, et seq.). The surface of a cell must have special properties; such sub- stances as shall lower the surface tension must collect there. These substances are probably lipoids (lecithin and Cholesterin) which have been demonstrated in every cell, animal as well as vegetable. The thickness of the transition layer varies, according to different ob- servers of different substances, from 1 to 25 /jl/jl; whereas the thick- ness of the material coherent pellicle (lecithin, etc.) does not need to be thicker than from 0.3 to 7 fxfjL. These are minimal figures which show that a membrane may be entirely invisible with the microscope and yet fulfil all the conditions of a true membrane as far as the transfer of material is concerned. . THE CELL 281 In brief our conclusions so far are: Every cell at its surface possesses a membrane which is dependent upon the composition of the interior of the cell. This membrane may be visible and may have been formed through the gelatinization of the cell protoplasm at the periphery. It may, on the other hand, be so thin as to be in- visible, being formed by the concentration and spreading out of such albuminous and fatty colloids as diminish the surface tension of the cell content at the interface. The cell membranes, developing as a result of the gelatinization of cell protoplasm, are at first, in youth, expansile and elastic; with increasing age these membrane colloids, depending upon their environment and upon chemical influences, or as a result of mere colloid aging phenomena, become poor in water and lose their elasticity. [In his " Growth and Form," Cambridge, 1917, D'Arcy W. Thomp- son invokes the aid of colloid phenomena in discussing the dynamics of cell Hfe. Tr.j CHAPTER XVII. THE MOVEMENTS OF ORGANISMS. The Movements of Lower Organisms. Freedom of the will is still a problem in philosophy, and even the investigation of the purely reflex phenomena and actions of higher organisms is still entirely in the stage of observation and measure- ment. In any case, it is still impossible to connect the external stimulus and the resultant action by a series of obvious physical and chemical processes. It is otherwise in the case of the movements of certain portions of plants and of the lowest organisms, especially certain amebse and their relatives, our symbiotic blood fellows, the leucocytes. In this case, opportunities are offered for an exact explanation of their movements and actions; but even here analogies must frequently carry us over gaps. It is customary to refer to such regulated movements of lower organisms and portions of plants as tropisms. [In this connection reference should be made to the discussion on "Animal Instincts and Tropisms in the Organism as a Whole," by Jacques Loeb. G. P. Putnam's Sons, 1916. John Hays Hammond, Jr., has constructed heliotropic machines which follow a lantern in the dark. The " retina " consists of selenium wire which changes its galvanic resist- ance when illuminated. Tr.] We speak of heliotropism when certain plankton organisms swim toward the light or when a tree or a flower grows toward the light. We speak of positive thermotropism if a root grows in the direction of a heat stimulus, of negative thermotropism when it grows away from it. Every fact in this connection is not only valuable in explaining the subject but serves as well to enrich the meaning of the term ''stimulus." "Stimulus" is an expression employed in biology wherever the more profound causes are not evident. Martin H. Fischer has already indicated how tropisms may be explained in analogy to curling sheets of gelatin. Th. Parodko contributed extremely valuable studies on plant tropisms. He stimulated growing roots from one side and they became crooked. The stimuli were chemicals, heat and traumata. He concluded that all these tropisms might be explained by protein 282 THE MOVEMENTS OF ORGANISMS 283 coagulation in the affected cells. All substances and concentrations which salt out or precipitate protein proved to be chemotropic. In connection with positive and negative chemotropism, salts of the alkalis and earth alkalis could be arranged in a lyotropic series similar to that we have repeatedly found in the precipitation of albumin and the swelling of gelatin, fibrin, etc. The salts of the heavy metals act still more strongly and always negatively chemotropic. Those movements which are manifest as general effects are still the most accessible to investigation. When placed between electrodes, bacteria, spermatozoa, yeast cells and red and white blood corpuscles migrate to the anode; amebse pass to the cathode. Although organized suspensions and colloids migrate either to the anode or a few {e.g., iron oxid hydrosol and aluminium oxid hydrosol) to the cathode, hydrophile organic colloids such as thoroughly dialyzed albumin and gelatin pass in no definite direction when placed in an electric field; they acquire a definite direction only by the addition of electrolytes. OH ions cause an anodal and H ions a cathodal migration. Since the organisms mentioned, considered as a whole, are hydrophile organic colloids, we must assume that their direction of migration in the electric field is determined by the ions clinging to them. Normal albumin with a content up to 0.01 normal NaHCOs still migrates to the anode. We need not be surprised, therefore, that the majority of micro- organs and microorganisms also migrate to the anode. The problem reduces itself to determining the direction taken by pure albumin. The cathodal migration of amebse is remarkable and requires more thorough study. Equally remarkable is the fact observed by H. Bechhold,* as well as by M. Neisser and U. Friedemann,* that agglutinated bacteria lose their direction of migration (p. 205), agglutinin having produced a neutralization. Following the ideas of G. Berthold,* we are nowadays tempted to explain by a simple formula certain individual movements of the lower organisms and of leucocytes; that is, by changes in surface tension.^ A fluid or semifluid structure which is constantly under the stress of surface tension assumes a spherical form, as, for instance, oil in a mixture of alcohol and water. If such a drop is placed between two other phases, a change in form occurs, and mth it a movement. A drop of oil on the surface of water spreads out; every moistening brings about an enlargement of the surface, a spreading out upon the moistened body (see p. 17). A structure may suffer a change of surface tension in some single spot, locally, so that a movement ^ A full bibliography is given in L. Rhumbler's " Zur Theorie der Oberflächen kräfte der Amöben": Zeitschr. f. wissensch. Zoologie, 83. 284 COLLOIDS IN BIOLOGY AND MEDICINE occurs; this may be induced for instance by an electric charge, chemical reactions and the like. Thus a structure may retain its general spherical form yet increase its surface at some single point, flattening out, putting out limbs, pulsating, making slight movements which may be explained purely by physical chemistry. The vital phenomena of amebse and of leucocytes which are evidenced espe- cially by movements of the plasma may be regarded as changes in surface tension. Portions of plasma (pseudopodia) are far ex- tended and the remainder of the body follows them, so that move- ments of progression arise. Sometimes the pseudopodia surround foreign bodies, a starch granule, a bacterium or the hke and draw it into the ameba or leucocyte; ingestion of food thus takes place. The migrations of an ameba, according to L. Rhumbler, may be deceptively imitated with a drop of chloroform in the following way: A Petri dish is covered with an alcoholic solution of shellac and the excess is poured off, so that after a few minutes the shellac layer is superficially hardened. Boiled water is then poured into the dish and a drop of chloroform dropped on the shellac with a pipette. Immediately the drop begins its characteristic migration, especially if it is pushed with a glass rod inserted between the chloroform and the shellac layer. The phenomenon is explained as follows : a marked surface tension develops between the chloroform, the water and the moist shellac layer; soon chloroform and shellac commence to be moistened at some point and at this point the surface tension of the chloroform is lowered and it seeks to spread itself out. In this way the chloroform drop progresses in a way similar to the flatten- ing of the advancing margin of an ameba. The thin shellac layer is dissolved by the chloroform flowing over it, so that the path traversed by the drop ''appears as if cut out of the shellac." Still more deceptive is the similarity of movement if one does not take a surface entirely covered with shellac, but prescribes the path of the drop by a fine shellac line and retards the movements by the ad- dition of Canada balsam or neat's-foot oil to the chloroform. Ac- cording to the proportions of chloroform, size of drop, thickness of the shellac layer and the degree of its dryness, the movements may imitate the most diverse kinds of amebse. If a drop of chloroform is placed on a spot of shellac which branches in various directions, an imitation of the spreading of pseudopodia is obtained. The tak- ing up of nourishment (taking up of oscillaria threads by ameba verrucosa) may, according to L. Rhumbler, be imitated when a drop of chloroform in water comes into contact with a thread of shellac; the drop completely envelops the thread of shellac and rolls it up into itself. THE MOVEMENTS OF ORGANISMS 285 At times small amebse are pursued by larger ones, the former change their direction and their speed, the pursuer continues its journey and catches its prey, which may again escape, and the pursuit continues. All these processes are explained according to L. Rhumbler, without invoking a conscious intelligence and pur- poseful movements, by the trail left beliind by the pursued ameba, just as the chloroform drop pursues the track of shellac mentioned above. Though the movements are so similar and the explanation by changing surface tension is so clear, we are still forced to enquire how the surface tension of amebse and leucocytes is changed. An- alogy is quite absent in the character of the substances whose sur- faces are in contact and in the physical process (solution of the shellac) that takes place. It was assumed that substances which diminish surface tension (for instance, soaps, albuminates, L. Michaelis) form at the point of motion and then break up again. Though a definite demonstration has not been possible, I shall discuss an hypothesis of L. Hirschfeld* which has much to recommend it in certain cases. We know that an electric charge depresses the surface tension (see p. 87) but the question is whether the development of an electrical charge at any point of a mass of protoplasm is conceivable. Let us consider the circumstances under which a bacterium approaches an ameba that puts out a Pseudopodium, envelops the bacterium and draws it in. Between two electrodes, amebse migrate to the cathode and bacteria to the anode. H ions diminish surface ten- sion, causing the extension of pseudopodia as demonstrated by the plentiful formation of pseudopodia upon fixation with osmic acid; OH ions cause an increase of surface tension and a retraction of pseudopodia. If we imagine a bacterium to be a negatively charged particle which gives off H ions, by dissociation it will lower the surface tension at the presenting point of the ameba and occasion the appear- ance of pseudopodia. When the bacterium is surrounded, there is an equalization of charge, the surface tension is raised and the pseudo- podium is retracted with the bacterium. L. Hirschfeld attributes the positive charge of amebse to the excretion of CO2. If the metabolism of the ameba is impaired, the formation of CO2, and with it the mobility of the ameba, are diminished. What occurs in the case of amebse may be applied to the special case of phagocytosis. It was the phenomena occurring in amebse that led Elie Metchni- KOFF to his fundamental studies on phagocytes, scavenger cells. Thus he names such white blood corpuscles as attack by taking up and digesting microorganisms entering the blood stream. They are the defending army of the organism, and according to E. Metchni- KOFF, the most important weapon in the fight against disease germs. 286 COLLOIDS IN BIOLOGY AND MEDICINE L. Hirschfeld is supported in his theory by the statement of H. Bechhold* that lactic acid (H ions) increase the phagocytic activ- ity of leucocytes, whereas alkalis (OH ions) are without such effect. This introduces us to one of the most important fields as yet [al- most] untouched by colloid investigation, Chemotaxis, the experimental study of which from modern viewpoints ought to prove most promising. In 1884, E. Stahl and de Bary on the one hand, and W. Pfeffer on the other, simultaneously gave their attention to the nature of Chemotaxis. They studied the lower monera, plasmodia of mjrxo- mycetes (slime fungi) bacteria, flagellates, wheel animalcules, the clustered spores of algae, the spores of ferns, mosses, etc. The essence of Chemotaxis lies in the attraction of these unicellular organisms by certain substances (positive Chemotaxis) and their repulsion by others (negative Chemotaxis), while other substances do not affect them at all. If for instance a cane sugar solution is placed in a very narrow test tube, and the open end is dipped into a drop of moss spores, the latter will pass into the tube, attracted by the cane sugar. It is necessary to assume some such chemotactic re- lation between eggs and spermatozoa, especially of aquatic animals, as the spermatazoa discharged into the water are attracted by the eggs. We owe our knowledge of the chemotactic action of leuco- cytes of the higher animals to C. A. Pekelharing and especially to Th. Leber who gives in his classical work, "Die Entstehung der Entzündung," a wealth of experiments in which the most varied substances were introduced into the eyes of rabbits. In the same field, but it is quite obvious independently, he was followed by Mas- sart and J. Bürdet. We reproduce the following series of substances with a chemotactic action (after A. Gabritschewsky) to show how difficult it is to ex- plain the existence of Chemotaxis on a single principle. Substances Showing Chemotaxis. Negative. Absent. Positive. 10% K and Na salts Distilled water 1% papayotin (for rabbits) l-io% glycerin Carmin powder Living and killed cultures of: Bile 0.1 to 1% K and Bacillus pyocyaneus 10% alcohol _ Na salts Bacillus prodigiosus Chloroform in aqueous Phenol Bacillus of anthrax solution 1% antipyrin Bacillus of typhoid 0.5% quinine solution 1% phloridzin Bacillus of hog erysipelas 0.1 to 10% lactic acid 1% papayotin (for All the bacteria that have Jequirity frogs) been studied excepting the Sterile culture of chicken 1% glycogen bacilli of chicken cholera cholera bacilli 1% peptone Bouillon Aqueous humor Blood THE MOVEMENTS OF ORGANISMS 287 It may be concluded ffbm this that a substance may be neutral for certain leucocytes and positively chemotactic for others (papayotin), and that the chemotactic relation may vary with concentration (K and Na salts; with reference to lactic acid and quinine solutions see pp. 286-288). What relation does all this bear to the theory of changing surface tension? Some data are in its favor. The very first observation of E. Stahl on the plasmodia of sethalium septicum of tanner's bark gives a decided impression that a surface phenomenon is involved. When he brought such a plasmocUum clinging to the internal surface of a glass in contact with pure water by introducmg the water from below, the Plasmodium spread out uniformly; if he introduced tannic acid, it trav- eled downwards; and on the addition of from 1/4 to 1/2 per cent sugar solution, it traveled upwards. It is just this action of tannic acid which tans the surface of protoplasmic mucus and the phenomenon of spreading out in pure water that point to surface forces. They are also suggested by the observation of Ranvier, according to whom leucocytes spread out more, the larger the surface development of the given body (better on rough than on smooth surfaces and especially well upon elder pith). On the other hand, we recognize from what has been said that the theory which attributes the decrease in surface tension to an electrical charge does not suffice for the ex- planation of all phenomena. An intensely positive chemotactic action is possessed not only by bacteria, but also by extracts and proteins obtained from them. The chemotactic experiments under- taken on the bodies of higher animals (eye, pleura, etc.) do not justify a physico-chemical explanation, because in this instance two factors coexist. The substance itself may act chemotactically; on the other hand, it may be inactive yet cause a necrosis of the adjoining tissue, which then becomes chemotactic and simulates activity on the part of the substances under investigation. [Else- where (p. 234) reference has been made to the observations of A. B. Macallum. His monograph " Surface Tension and Vital Phenomena," No. 8 Physiological Series, University of Toronto Studies, 1912, includes a bibliography. Tr.] Possibly the very original "Quantitative Studies on Phagocytosis" of H. J. Hamburger and Hekma* will permit conclusions concern- ing the causes of the protoplasmic movements of leucocytes, when a method shall have been discovered for measuring the surface tension of protoplasm against water and salt solution. Even now it may be recognized from these studies that the causes of movement are quite complicated since it has been shown that the calcium ion has an entirely specific action in stimulating phagocytosis. If such action 288 COLLOIDS IN BIOLOGY AND MEDICINE were due merely to the electric charge possessed by Ca as a divalent ion, we would expect the same effect from barium, strontium and magnesium; this however is not the case. Especially noteworthy is the fact, only recently studied by G. Denys and Leclef, Weight and his pupils, and Neufeld among others, that leucocytes are stimulated to the phagocytosis of certain bacteria only by the presence of serum, and that, on the one hand, the intensity of the phagocytosis is dependent upon the virulence of the bacteria, and, on the other, upon certain properties of the serum, closely related to those which determine immunity. To the colloid chemist, it is of importance to determine whether the general colloid properties of serum play a role in phagocytosis, and whether the serum may be replaced by other colloids. H. Bechhold* showed that egg albumen, which stands nearest to serum in respect to its colloid properties, caused no phagocytosis, whereas Witte's peptone, a markedly broken down protein, has such an action. In the case of Chemotaxis, as in the case of phagocytosis under the influence of opsonins (or certain hypothetical irritants which increase the appetite of leucocytes) only comprehensive quantitative experiments will yield material utilizable for the development of a physico-chemical theory by the colloid chemist. Although, for instance, quinine is regarded as a substance which inhibits phago- cytosis, M. Neisser and Guerrini * have shown that in minimal doses it increased the appetite of leucocytes. It may be said in conclusion that the surface tension of leucocytes in relation to the surrounding medium (serum) must be very low. On page 16, we saw what force is necessary to change the form of such small bodies (leucocytes have an average diameter of from 6 to 8 /x). If we recall what changes in surface tension a leucocyte may undergo in phagocytosis, and the very great changes in shape suffered in traversing the tissues, we are forced to ascribe to them a very low surface tension, much lower than that possessed, e.g., by red blood corpuscles. The Movements of Higher Organisms. The movements of higher organisms are controlled by the nerves and accomplished by the muscles. In the present state of our knowledge and in the limits of this book we can only consider this question: From what physical and chemical processes does muscle contraction result? For this purpose we shall first consider the muscle as a colloid system and endeavor to gain an idea how a contraction occurs. THE, MOVEMENTS OF ORGANISMS 289 Muscle as a Colloid System. In the case of higher mammals, muscles constitute approximately 43 per cent of the entire body. Since they have a greater range of swelling than all the other organs (see p. 219), besides their usual function as a water reservoir, they are of great importance. As regards swelling, they behave very much like fibrin or gelatin. It was formerly believed that the circumstances of sweUing in muscle, which were at first chiefly studied in the case of frog muscle, could be explained by osmosis, but the quantitative studies of J. Loeb,* followed later by A. Durig, C. E. Overton * and R. W. Webster, showed that no satisfactory solution could be thus obtained. If the osmotic conditions alone were determinative, the muscle should retain its water in isotonic solutions, shrinking in hypertonic and swelling in hypotonic solutions. But this is not by any means the case, since there is a material difference between solutions of electro- lytes and of nonelectrolytes. Whereas neutral salts greatty diminish the swelling produced by acids and alkalis, this property is not pos- sessed by nonelectrolytes (cane sugar, ethyl alcohol, methyl alcohol, urea and glycerin). Even the supposition of a lipoid membrane does not explain the phenomena, since cane sugar is as insoluble in lipoids as are most of the neutral salts. As early as 1901, A. Durig concluded from his investigations with whole frogs that the laws which are invoked in osmotic processes alone are inadequate; in this case muscles are chiefly concerned in the absorption and relinquishment of water. Martin H. Fischer* was the first to direct attention to the fact that for dead muscle, qualitatively and to some extent quantitatively, similar laws gov- erned the taking up and the relinquishment of water as governed unorganized colloids capable of swelling. To summarize his results briefly: muscles swell more in acids and in alkalis than in water, and indeed, in hydrochloric acid, nitric acid > acetic acid > sulphuric acid. The maximum amount of water that a muscle can absorb under the circumstances is about 246 per cent of the original muscle weight, or 13 times the dried muscle substance. It therefore possesses, it is true, a smaller swelling capacity than gelatin which can take up from 15 to 25 times, or fibrin which takes up upon solution 30 to 40 times, its dried weight. The absorption and relinquishment of water by muscle is a re- versible process, yet M. H. Fischer emphasizes the fact that during the time of his experiments no complete reversibility was observed, that "every change of condition left its permanent results." 290 COLLOIDS IN BIOLOGY AND MEDICINE Salts diminish the swelhng of muscle in acids and alkalis in a way similar to the case of fibrin and gelatin, though not so obviously. There is, indeed, a very important difference between dead and living muscle: the swelling of dead muscle in distilled water, for instance, is brought about by the formation of lactic acid, which sets in within a few minutes. If this were not the case, a living frog would swell up as much in fresh water as a dead one.'^ According to M. H. Fischer, a dead muscle retains its form in a 0.7 per cent NaCl solu- tion, not because the same osmotic pressure exists inside and outside the cell, but because the concentration of the NaCl solution is just sufficient to overcome the action of the acids formed in the excised muscle. We must again point out here that the experiments of W. BiLTZ and A. von Vegesack * show that if colloids are present in a medium, the presence of isotonicity does not by any means permit us to infer that equal osmotic pressures exist. Against M. H. Fischer's experiments, the objection has been raised that dead muscle possesses no semipermeable membrane, so that its swelling follows laws similar to those of fibrin, etc. In living muscle, however, semipermeability exists; on this account the re- sults of M. H. Fischer cannot be transferred to living muscle. There are also certain discrepancies in respect to some nonelectro- lytes; thus, for instance, dead muscle does not swell up in isotonic sugar solution; this does not accord with Fischer's theory. [Sugar has a specific dehydrating action. Tr.] The studies of E. B. Meigs * have illuminated these discrepancies; they showed a definite difference between smooth and striated muscles. Smooth muscles are involuntary and occur in automatically acting organs (intestines, urinary bladder, iris, etc.), and especially widely distributed among the lower animals. They contract much more slowly than striated muscles. E. B. Meigs concludes that smooth muscle is not surrounded by a semipermeable membrane, in other words, osmosis is not a factor, but that they behave toward electro- lytes like any hydrophile colloid, fibrin or gelatin, with reference to change in volume. The behavior of striated muscles is quite different. To understand it we must briefly recall their histology. Muscles consist of bundles of fibrils, longitudinal fibers which are surrounded by a connective tissue sheath. Each fibril, that is, every minute fiber, is surrounded by a membrane, the sarcolemma, and is bathed in a fluid substance, the sarcoplasm. The individual fibrils are striated at right angles to their axes. The striations appear microscropically as alternating dark and bright zones; while the latter are isotropic, the dark striations are doubly diffractive, anisotropic (see Fig. 49). 1 [If the circulation of a living frog is impeded so that local acidosis develops, local swelling also develops. Tr.] THE MOVEMENTS OF ORGANISMS 291 Fig. 49. Striated muscle fiber. (Stöhr.) E, B. Meigs * studied the rate at which fresh muscles in- creased their weight in water and in salt solutions. He concluded from his study that the weight increase is the result of two processes : At first, water is osmotically taken up by the sarcoplasm of the fresh (still irritable) muscle; after the muscle is dead, lactic acid forms, the semipermeable membrane of the fibrils (the sarcolemma) becomes permeable and now the fibrils swell up at the ex- pense of the sarcoplasm fluid and are thus 'short- ened; this is evidenced by rigor mortis (0. von FÜRTH and Lenk). The proteins become co- agulated through the accumulation of acid; this especially induces a shrinking and thus a relaxa- tion of rigor mortis. By this experimentally es- tabHshed explanation 0. von Fürth and Lenk have cleared away an old fallacy that the onset of coagulation induced rigor mortis. By artifi- cial fatigue {e.g., electrical stimulation of an excised frog's muscle) the accumulation of acid and the consequent swelling of muscle in dilute salt solution is much hastened (C. Schwarz *). It is a well-kno^vn fact, moreover, that after great muscular exertion (forced marches, convulsions, hunted prey), rigor mortis sets in sooner than when death overtakes a rested organism. When rigor mortis disappears striated muscle behaves like an hydrophile colloid, whose swelling and shrinking are unhindered by semipermeable membranes. A further study of E. B. Meigs* is concerned with the nature of the semipermeable membrane of a fibril. It tends to show that the latter consists of calcium phosphate. Collodion membranes impreg- nated with calcium phosphate proved impermeable for salts, sugar and amino acids, but were somewhat permeable for glycerin and urea and easily permeable for ethyl alcohol. They were moderately perme- able for potassium chlorid as was to be expected. The predication of a semipermeable layer of calcium phosphate explains two facts very well: 1. The suspension of the semipermeabiHty of muscle after death (the accumulation of lactic acid destroys the membranes) and 2. the importance of calcium for the maintenance of semipermea- biHty in living muscle; since the layer of calcium phosphate is de- stroyed in a neutral lime-free solution. A unique observation was made by M. H. Fischer and P. Jensen * upon the water in muscle. They put the gastrocnemius of frogs into narrow glass tubes, cooled them do\^^l to —76° in a mixture of ether and solid CO2, and followed the curve of cooling wdth a needle- 292 COLLOIDS IN BIOLOGY AND MEDICINE shaped thermocouple. With a muscle of average size, the phenomena are about as follows: Within 3 or 4 minutes there is a very slight cooling; within another 8 or 10 minutes the water freezes in the muscle and further cooling occurs. This part of the curve is quite characteristic for the fixation of water. It should fall more steeply than the curve of control water or of physiological salt solution. If it is less steep, it is "a sign that there is some process in the colloid structures which hberates heat." ^ In H. W. Fischer's and P. Jen- sen's investigations, it was shown that it is necessary to distinguish two kinds of water fixation in freezing muscle. ''After or during death by freezing, there occurs a phenomenon by which water is fixed in some unknown way and by which it is again liberated at lower tem- peratures," and indeed the amount of water fixed in a muscle in- creases with the amount of disturbance (whether frozen once, twice, heated to 100° or boiled). In this case, also, it is seen that two kinds of water fixation exist. The relation between cooling and the death of muscle by freezing is very interesting. The degree of "vitality^' was measured by the lifting capacity of a muscle in response to stimulation. It was shown that cooling the muscle to the point of freezing, and even freezing out the water to a certain extent, did no harm, but if the muscle was cooled 1.5° C. more, it died. The inferior thermal margin between life and death of muscle is, therefore, only 1.5° C. wide. The normal state of swelling in muscle is conditioned by a normal content of electrolyies. This may vary greatly for different classes of animals; for instance, according to J. Katz, the striated muscle fibers of the dog contain 3.5 times as much K, and those of the pike 14 times as much K, as Na. For the same species it appears to be uniform at the same age. A remarkable fact regarding muscles is their high potassium content, which is closely associated with their capacity to functionate. [See Macallum, also Burridge. Tr.] For a normal swelUng, the iso- tonicity of the surrounding solution appears to be of much less importance than a definite electrolyte content. This follows from experiments of E. M. Widmark,* according to which even 10 milli- mols CaCl2 in the surrounding solution produced a loss of weight amounting to 36 per cent (in the spUt muscle fibers). Muscle Function. Every stimulus, whether of thermal, mechanical, electrical or chemical nature causes an irritation in the living muscle which is manifested by a contraction. R. Höber,* whose investigations we ^ The mathematical basia for this is given in the original work. THE MOVEMENTS OF ORGANISMS 293 shall now consider, is primarily responsible for the electrochemical theory of this irritability. For our consideration, two electrical phenomena of muscle are important: In activity, that is during the contraction of muscle, electric currents (action currents) develop; the stimulated point in the muscle becomes negative in relation to the remaining fibers which are at rest. The same thing holds true for nerves in which no external sign of activity is discoverable. If, in an excised mollusc muscle, an injured point is united to an uninjured point of the mantle by a wire having a galvanometer in its circuit, the cut surface is negative and the mantle surface is posi- tive. The same electrical phenomena are observed in a resting nerve. This is called the current of rest, or, according to H. Herman, the de- marcation current (Herman calls the demarcation surfaces the interface between the injured, dead, and the uninjured, Hving, substance). Evidently action current and current of rest are due to the same cause. In his textbook, R. A. A. Tigerstedt states the phenomenon as follows : In muscle as in nerve, a stimulated point, or one which is injured in any way, is negative electrically to every other point which at that time happens to be at rest or uninjured. Let us consider how we may explain the direction and magnitude of different potentials which occur when muscle contracts. Elec- trical differences in potential arise on every interface between an electrolyte and a pure solvent or one containing less electrolytes. The simplest case is when an acid, e.g., HCl, is Hmited by pure water — then the more mobile positive H ion will rapidly advance and give a positive charge to the water while the acid is negatively charged by the more slowly moving negative CI ion. This applies to muscle, for lactic acid arises at the point stimulated or injured. The electromotive forces which are derived from a circuit of acids and water or crystalloid electrolytes are much smaller than we observe in muscle. Wo. Pauli invokes the colloidal properties of the protein ions in explaining the high electric tension which we obtain in muscle or even in the electric organ of the torpedo. Protein in general contains an amino acid with many NH2 and COOH groups. Let us illustrate the development of electromotive forces by the following diagram in which R represents the protein radicle and L the lactic acid radicle: OHCO . . NH2 + LH OHCO • • NH3 L OHCO. |< .NH2 + LH = 0HC0. P-NHg-fL OHCO- .NH2 + LH OHCO. .NH3 L 294 COLLOIDS IN BIOLOGY AND MEDICINE The difficultly mobile colloidal acid-protein ion immediately becomes positively charged at the surface of a neutral medium, and should it touch an acid medium its positive charge is raised and at the same time the acid field becomes more negative as the following diagram indicates : OH . CO ■ .NH2 OH . CO . Tl'T% OH . CO . R . NH2 + LH = + = OH . CO ■ |< • NH2 H + L OH . CO . -ML.mJ .NH2 OH . CO ■ . . NH2 for the H ion moves faster than the L ion. Measurements of series consisting of acids and acid albumin couples yielded potentials quite large enough to account for action currents. The development of such diffusion potentials in muscle would not be possible if the fibrils were not quite poor in salt and the sarcoplasm quite rich in salt. Since both the fluid and the fibrillar portions contain protein (see Bottozzi and his school) a couple consisting of acid albumin/acid/acid albumin yields no current. The current is reestablished through electrolytic dissociation of the acid albumin due to the salt in the sarcoplasm (see p. 292). If such couples are placed in series considerable electric tension (voltage) may be ob- tained. These results are in agreement with the fact that the normal properties of muscle are conditioned by definite states of swelling and electrolyte content. If frogs' muscles are placed in an isotonic solution of cane sugar or other nonelectrolyte (mannit, asparagin, etc.), they lose their irritabihty (e.g., for the induced current) but retain their volume; they do not swell as in distilled water in which the irritability is likewise suspended. The ability to contract is restored by Na ions (about 0.07 per cent NaCl) (C. E. Overton) as well as by Li ions; but it is not restored at all by K ions. The irritability is also sus- pended by isotonic potassium and rubidium salts. If the anions and cations are arranged in accordance to the extent with which they interfere with irritability, we obtain lyotropic series similar to those which we discovered for the salting out of colloids (see pp. 80 to 83); according to R. Höber,* C. E. Overton* and Schwarz,* they are as follows: inhibitory: K > Rb > Cs > Na, Li inhibitory: tartrate, SO4 > acetate > CI > Br,N03 > I > SCN. If an uninjured frog's muscle is dipped into an isotonic solution of a neutral salt and the part so treated is united with another part of THE MOVEMENTS OF ORGANISMS 295 muscle l?y a wire, we obtain a current of rest whose strength and direc- tion depends on the nature of the neutral salt. [The study of these cur- rents of action in the heart muscle has been elaborated into the science of electrocardiography, I know of no attempt to associate electro- cardiographic curves with changes in the colloids of the heart muscle in response to salts. Tr.] If the anions and cations are arranged according to their action on this current of rest (see R. Höber and Waldenberg *), we obtain series similar to the above. Since we have previously seen that the salting out of protein, the swelling and shrinking of gelatin and fibrin (which means the ionization) occur in similar lyotropic series, R. Höber concludes that the normal irrita- bility of muscle is dependent upon a definite condition of solution or swelling of its protoplasmic colloids; increased solution or precipitation of the colloids leads to loss of irritability. J. Loeb and R. Beutner are of the opinion that the current of inactive muscle due to salt (as well as the currents rising in plants because of an injury to some part) bears no direct relation to the condition of swelling of the plasma colloids,^ but is due to a hpoid membrane on the surface of the muscle or its constituent elements. The variation in activity of the salts chosen (NaCl, KCl, etc.) is due to their different threshold of solubility in the hpoid membrane. R. HÖBER correctly emphasized that for such questions of physio- logical function we need consider only those influences which are reversible. Substances causing a more or less irreversible change by means of aromatic anions require no further consideration here. The dependence of the irritab hty of muscle upon, and its relation to, the condition of the organ colloids are not unique. Examples of other organ functions were studied by R, S. Lillie *^ (movement of the cilia of the larvae of marine annehds) and by R. Höber *^ (the movement of the cihated epithelium of the frog) . The movements of cilia above mentioned cease upon the addition of various salts: in fact, of the alkali salts, Li salts are the most harmful. In hemolysis and in the diminution of the movement of ciha, the anion series shows an order the reverse of that for the diminution of muscle irritability, which means that the swelling of blood corpuscles and muscle are affected in an opposite way. Such well-known hemolytic agents as saponin, solanin, taurocholic acid, glycochohc acid and sodium oleate diminish the irritability of muscle in an irreversible manner; they evidently damage the lipoid plasma pelhcle (R. Höber *ii). 1 We must forego further discussion of the extremely interesting results of J. Loeb and R. Beutner since they have no direct bearing on colloids. 296 COLLOIDS IN BIOLOGY AND MEDICINE The accompanying table (in part after R. Höber) gives at a glance the action of the various alkaline salts, and parallel with it the extent to which such salts salt out hydrophile colloids. The question now arises, What are the colloid-chemical changes which occur as the result of stimulation and bring about the change in the shape of the muscle f We know from the investigations of G. Jappelli and D'Errico as well as of G. Buglia, that muscle absorbs water when it contracts (fatigue). This is not surprising, since acids are formed which favor swelhng (see p. 267). According to the conception of E. Pribram, the formation of acids and the contraction of muscle are closely associated. Even Th. W. Engelmann had already drawn the conclusion that during contraction, water passes from the iso- tropic water-rich layer of the striated muscle into the anisotropic water-poor layer, which swells. This is due to the transfer of acids from the sarcoplasm where lactic acid is created by stimulation. Water flows from the blood and the lymph into the isotropic layer, so that as a result of the contraction, the entire muscle is richer in water. We must picture of the shifting of fluid within the fibrils as occurring in such a way that the anisotropic layer, which, according to MUNCH, is spirally arranged, can expand only from side to side when it swells (at the expense of the isotropic layer). This causes a trans- verse thickening of the muscle fibers and a shortening in length, a contraction. If the lactic acid in the living muscle is consumed or otherwise neutralized the process is reversed and the muscle regains repose. Streitmann and M. H. Fischer constructed from catgut a working model of muscular contraction. The catgut strands represented the anisotropic substance and the sarcoplasm was replaced by water, acids, and salt solutions. For the sake of completeness, we shall refer to one other theory which is by no means as well estabhshed experimentally as the one described. Bernstein first suggested the idea that muscular con- traction was associated with changes in surface tension. As has been mentioned previously, muscle is characterized by an especially high content of potassium. From the researches of A. B. Macallum we are compelled to assume that it has a most important function dur- ing contraction. In contractile tissues (muscles of frogs, lobsters, beetles, etc.), according to A. B. Macallum * and his pupils, the potassium seems to be localized in the dark zones of the resting muscle fibrils, especially at their surfaces. From this, A. B. Macallum concludes that the surface tension must be lowered in these zones. With the contrac- THE MOVEMENTS OF ORGANISMS 297 o CO S Ü V A I— I "* Ü ^ O V S ^ o V o ^ Ü ^ tn" V tn" m I m V « A Ü . Ü GO t-i 32 o CO Ü A 6 A CQ O CO A Ü A d A .„ h^ h^ ^^ A V 1^ ^ 1^ ^ ^ ^ AAV !ZJ CC tß Ü Ü Ü V 72 o V A A of A O Ü A Ph Ph rt AAV >v< iv< k> V A X Xi Xi P^ Ph rt V A 03 r^ o o u Ö o Ö o O 73 ^ -t^ 73 S 3 T3 "3 c3 ^ ^ ^ •§ 03 « Q Ph O 02 A -1^ o3 c3 u CO V o q; o 03 o CO Ü A V /\ V pq Ü U o A V o ^ 1— ( o ^ Ih" A ^ A w ^ V >^ V U Jh C) 1— 1 CO P4 02 1-^ 1-^ A V A V A V A V ^ :s S .5 -S Ö c3 A V A . O P3 ^ V A ^ m m A V c3 13 W) S --i -2 O O O O Ö Ö Ö fl _o _o _o o ■+^ '-to '+i '-5 o3 o3 c3 o3 SJ -to -(J +J _'&, ]a 3 |a, o 'o 'o 'o 03 4,000 > 4,000 Vorticella. 170,000 92,000 36,000 9,500 5,200 > 4,000 > 400 It is noteworthy that the lethal threshold for salts of the same metals are very similar for the same dilution and for the same con- tent of metal. Colloidal silver has absolutely no effect on moulds. I found that a 1 per cent collargol solution which had been loft unstoppered was covered after a time with a species of mould. Similar observa- tions were made by Filippi * with penicillium and aspergillus in the case of different colloid metals. R. Zsigmondy*^ mentions that moulds grew on his gold hydrosol and that the solutions were grad- 368 COLLOIDS IN BIOLOGY AND MEDICINE ually decolorized by them as the gold precipitated on the mycelia and stained them black. Earher investigators (Crede, Cohn, Brunner, Netter) ob- served only a moderate inhibition of growth (1 : 2000 to 1 : 6000 in the case of staphylococcus aureus) but no destruction of the germs by colloidal silver. Recent studies of Cernovodeanu and V. Henri * on anthrax bacilli, B. coli, staphylococcus pyogenes aureus and albus, B. dysenteria, etc., show a strong bactericidal action of silver hydrosol in test tubes; researches of Charrin, V. Henri and Mon- NiER-ViNARD * show the Same effect in the case of B. pyocyaneus. The size of the particles in a hydrosol is of very great importance, and in fact the finely granular red solutions are much more active than the coarser green ones; the former completely inhibited growth in dilutions of 1 : 50,000 to 1 : 100,000. [Jerome Alexander has produced especially fine dispersion by a new principle. Tr.] Similar results were obtained for pneumococci by Chirie and Monnier Vinard.* According to G. Stodel,* colloidal mercury in a dilution 1 : 132,000 inhibits the development of B. typhi and of staphylococci. On account of the results obtained with colloidal silver,^ as well as because of the lack of irritating effect and of toxicity (it was pos- sible to employ it in large doses subcutaneously and intravenously), the hopes for its therapeutic action were justified. It is remarkable that, instead of extensive especially planned animal experiments, clinical experiments which were at times favorable and at times un- favorable have occupied the stage. The number of times it has been employed clinically compared with animal experiments is com- paratively small, and it was tried on many hopeless cases. The judgment of the results depends largely on the experience of the chnician and is much influenced by the subjects; in short, the results hitherto obtained lead to nothing definite. On this account the indications for use are very inadequate. It is from the above- mentioned exhaustive researches of M. Ascoli and G. Izar * that an idea of the mechanism of the action of metal hydrosols has been obtained. [Harry Culver (Jour. Lab. & Clin. Med., May, 1918) found that the gonococcidal action of colloidal silver (argyrol, pro- targol, silvol and nargol) was diminished in vitro by aging the solu- tion by light and by heat. He also found that the gonococci became resistant or adapted to -a particular preparation by growth in its presence. This was not a resistance to the other colloidal silver preparations but specific. The importance of the " protecting " sub- stance is evident from this experiment. Tr.] 1 According to Stodel also, colloidal mercury is less toxic than mercury salts. TOXICOLOGY AND PHARMACOLOGY 369 Ferments. Ferments are much reduced in activity by the salts of heavy metals. Since a parallelism has been shown to exist between the toxicity of colloidal metals and that of their salts, it was expected that the colloidal metals would exert a powerful action on ferments. It is a remarkable fact that the colloidal metals proved to be more or less indifferent. The digestion of albumin by pepsin, the digestion of gelatin by trypsin, the coagulation of milk by rennin, the cleavage of fat by pancreatic steapsin and lipase, the fluidification of starch by pan- creatin and takadiastase were uninfluenced by colloidal silver (see M. AscoLi and G. Izar*). L. PiNcussoHN * examined the following substances for their in- fluence on digestion with pepsin: chemically prepared hydrosols of silver, selenium, gold, copper, bismuth, mercury (Hyrgolum) and arsenic; and electrically pulverized preparations of silver, gold, platinum, mercury and bismuth. In no case was the activity of pepsin increased, but it was diminished by large doses, and least in the case of hydrosols obtained by electrical pulverization. E. FiLipPi * was unable to obtain any effect with colloidal metals (Au, Hg, Cu, Ni, Pd) upon fermentation in the case of yeast, pepsin, trypsin or rennin. SmaU quantities of silver hydrosols, on the contrary, activate the diastatic ferment of the liver and of the blood serum. According to H. J. Hamburger, the action of staphylolysin, the hemolytic excretion of staphylococci, is inhibited by collargol. Ac- cording to W. Weichardt, colloidal platinum and palladium neutral- ize fatigue poisons. In vitro, C. Foa and A. Aggazzotti were unable to demonstrate any action of silver hydrosol upon toxins, but they could if it was in- jected into the circulation immediately after the toxin. O. Gros and J. M. O'Connor obtained divergent results for the decrease in the strength of tetanus and diphtheria toxin produced by collargol. Autolysis. In marked contradiction to the inactivity of silver hydrosol on most ferments is the very considerable influence of metal hydrosols on the enzymes of autolysis. If any organ, the stomach, liver, spleen, etc., is kept, especially if kept at body temperature, changes occur in it which finally lead to a softening and decomposition characterized by a more or less extensive cleavage of the albumins, nuclcins, etc., involved. This decomposition occurs even though the organ is absolutely 370 COLLOIDS IN BIOLOGY AND MEDICINE sterile, so that incidental bacterial growths are not the cause; it is brought on by a series of different enzymes each of which has a definite function, and the process is called autolysis or autodigestion. All the hydrosols investigated, namely, those of silver, gold, plat- inum, mercury, palladium, iridium, copper, lead, ferric hydroxid and aluminium hydroxid have the ability to assist autolysis; M. AscoLi and his coworkers, by separately investigating the resulting products, were able to determine the action of the individual enzymes. For instance, the liver of a recently killed animal was cut up into small pieces and passed through a sieve; it was then diluted with water and distributed in a number of sterile vessels with 1 per cent toluol to pre- vent putrefaction. In one sample the albumins were immediately coagulated, and the total nitrogen, as well as the individual nitrogen fractions, determined. Varying quantities of metal hydrosol were added to the remaining vessels and they were kept for 72 hours at 37° C. Each portion was then tested for 1. Total nitrogen (according to Kjeldahl). 2. Nitrogen (as monamino acids). 3. Purin-bases (according to Salkowski). 4. Albumose-nitrogen (according to Baumann and Bömer). The difference between the total nitrogen and the sum of the other values gave the quantity of nitrogen present as diamino acids, peptone and ammonia. In general, there is an accelerating action on the total autolysis as well as on the cleavage of the nucleins, and the formation of monamino acids; though there are considerable quantitative dif- ferences between the different hydrosols. For instance, minimal quantities of Ir, Hg, Cu and Ag favor the autolytic process in general, yet decidedly larger quantities of Pb, Au, Pt and Pd are required for this purpose. The same facts hold for the formation of monamino acids. Small doses increase, while larger quantities of hydrosols interfere with the cleavage of nucleins; however this does not hold true for silver, platinum and gold hydrosols. Under ordinary circumstances the uric acid formed during autolysis is broken down still further by a uricolytic ferment; the action of this ferment is inhibited by silver hydrosol. Though there is no difference between the action of stabilized and unstahilized silver upon autolysis, such a difference was noticeable after the addition of defibrinated blood. Defibrinated blood interferes with the acceleration of autolysis due to unstahilized silver hydrosol, but it does not do so in the case of the stabilized hydrosol. This observation is also of great interest in connection with the TOXICOLOGY AND PHARMACOLOGY 371 theory of protective colloids. A priori we would be justified in he- lieving that no diflference exists between stabihzed and unstabilized metal hydrosol, but that a stabilization could be produced by the dissolved albumins of the hashed organ or of the added blood. The above example indicates the delicate adjustments in the mechanism of colloid protection. [Different substances may compete for the protector, thus establishing " preferential " protection. Tr.] It is interesting to note in addition, that the above investigators found that minimal traces of prussic acid, mercuric chlorid and Cyanid, arsenious acid and carbonic oxid had as toxic an effect on the autolytic action of silver hydrosol as upon its abihty to split hy- drogen peroxide. This process which was exhaustively studied by G. Bredig may be made to regress so that the metal hydrosols may "recover." The identical observation was made by M. Ascoli and G. IzAR in respect to the autolysis by poisoned silver hydrosol. Blood: Hydrosols of silver, lead and mercury have the abilit}^ to dissolve red blood corpuscles, whether the hydrosols are stabilized by gelatin or not (M. Ascoli *^). It is also interesting to learn that pure powdered silver causes hemolysis, though this proceeds very slowly. The same silver powder when repeatedly used for hemolysis be- comes inactive; serum inhibits hemolysis by silver. H. Bechhold ^ observed that a drop of mercury causes strong hemolysis, which serum did not inhibit. He also observed hemolysis with metallic lead, though this was much weaker than in the case of mercury. Metallic copper hardens the erythrocytes. Poisons do not interfere with the action of silver hydrosol. It is necessary in these effects to distinguish between the specific activity of the metal involved and the generic activity due to the development of surface. Hemolysis is induced by quite indifferent suspensions, by kaolin (Friedberger and his pupils) as well as by barium sulphate and calcium fluorid (0. Gengou). Such hemolysis is inhibited by serum. After AcHARD and E. Weill, as well as A. Robin and E. Weill, had studied the influence of colloidal silver, and G. Stodel- had studied the influence of colloidal mercury upon erythrocyte pro- duction, E. FiLiPPi, and later Le Fevre de Arric, carried these in- vestigations further and extended them to other metal hydrosols. The results in brief show that the red blood corpuscles are at first diminished to a greater extent than the white. Later there is a con- siderable increase of both red and white blood corpuscles. After the 1 As yet unpublished. 2 The fact that G. Stodel did not observe hemolysis of dog's blood with electri- cally pulverized colloidal mercury is remarkable, and deserves further investigation. 372 COLLOIDS IN BIOLOGY AND MEDICINE prolonged injection of hydrosols the red blood corpuscles and the hemoglobin are somewhat increased, but there is no noticeable in- crease of leucocytes. Silver, copper, manganese and mercury prove most active; platinum, palladium, gold and nickel are much weaker. Identical results are obtained with small doses of the salts of these metals. This does not completely accord with the results of 0. Gros and J. M. O'Connor,* who observed an immediate increase of the polynu- clear leucocytes just as occurs after the introduction of any other foreign substances. Very noteworthy is the observation of Filippi, that colloidal silver, copper and mercury introduced into the circulation markedly in- crease phagocytosis. The following table obtained with slightly different experimental conditions on rabbits illustrates this: Phagocytosis of Aleuron and Carmine. Normal. Ag. Cu. Hg. Pt. Per cent. 3.12 5.20 Per cent. 27.50 37.80 Per cent. 17.80 40.16 Per cent. 38.00 16.10 Per cent. 8^20 Le Fevre de Arric found, on the contrary, that this assumption could not be generalized. In experiments with silver hydrosol (electrargol) he found in guinea-pigs an increase in the phagocytic activity for colon and typhoid bacilli; in rabbits there was a diminu- tion for typhoid bacilli. In both guinea-pigs and rabbits there was an unfavorable effect on the phagocytosis of pyocyaneus and staphylo- cocci. Metabolism. Naturally, the processes occurring in the living organism are far more complicated than in the individual organ elements or in the dead organ. However, since there were obtained from the study of autolysis viewpoints for the action of hydrosols on the disintegration of nitrogenous constituents, the investigation of the nitrogen change in the living organism offered a prospect of profitable study (M. AscoLi* and G. Izar,*^ Filippi and Rodolico). For this purpose bitches were fed entirely on bread made from wheat or rye flour. The total nitrogen in the feces was determined, and in the urine the total nitrogen, the urea nitrogen and the uric acid. In a previous series of experiments with men, like determinations were made (excepting of the N of the feces) as in the experiments under- taken on rabbits by E. Filippi and Rodolico. Metal hydrosols were administered intravenously. The results were concordant. TOXICOLOGY AND PHARMACOLOGY 373 The result of the experiment was as follows: unstabilized silver hydrosol (prepared according to G. Bredig) as well as collargol had no action in small doses. Silver hydrosol (prepared according to G. Bredig), stabilized with gelatin, increased the nitrogen metabolism; the nuclein metabohsm was chiefly affected since there resulted a decided increase in the elimination of uric acid in the urine. Sil- ver hydrosol stabilized by gelatin has a more powerful action than the corresponding quantity of silver nitrate, silver thiosulphate or silver albuminate, which exert a qualitatively analogous action. On the other hand, the N elimination in the feces is decreased. Mercury and lead hydrosols have a similar effect, differing only in the time curve. Large quantities of collargol also increase the uric acid excretion. Temperature Curve. The injection of a few cubic centimeters of silver hydrosol causes a rise of temperature of varying but usually brief duration (M. AscoLi and G. Izar*); on the other hand, the unstabiUzed hydrosols have no observable effect on temperature (Bourgougnon*). This corresponds with the observations on autolysis described above. Distribution. Finally, we must inquire, what becomes of the injected silver hydrosol. This has already been investigated, at least as far as con- cerns collargol injected intravenously. G. Patin and L. Roblin * found it chiefly in the liver but to a less extent in the kidney. They contend that there occurs a concentration and gradual excretion through the kidneys. S. Bondi and A. Neumann showed that col- largol as well as other indifferent suspensions (India ink, fat) dis- appear from the circulation within 1/2 to 1 hour after intravenous injection and are temporarily deposited ift the liver, bone marrow and spleen. It is the star cells of von Kupffer which chiefly take up these suspensions. J. Voigt contributed especially accurate researches. He traced the fate of the stored silver in the more important organs bj^ examin- ing microscopic sections in the ultramicroscope. Of his findings let us emphasize particularly that it made a difference in the distribution of the silver in the individual organs whether the animal was over- whelmed by a single large quantity of silver solution or smaller repeated doses were injected. There were definite differences in the pictures obtained with different colloidal metals and metallic com- pounds. According to personal, hitherto unpublished communica- 374 COLLOIDS IN BIOLOGY AND MEDICINE tions from J. Voigt the silver is precipitated at the site of injection after intramuscular injections and in the peritoneum after intra- peritoneal injections, whence it is gradually transported to the internal organs. It is still an open question whether the transportation is purely mechanical or results from solution and reprecipitation. Therapeutics. It is obvious from the preceding statements that metal hydrosols may, from very different causes, exert a therapeutic action. In in- fectious processes we may imagine that there is a direct action on the excitants of infection; although this may be due to an indirect action inasmuch as the hydrosol stimulates the formation of anti- bodies and phagocytosis, or it may injure the infecting organisms by intensifying metabolic changes in some way. In view of G. Bredig's experiments on the catalytic action of col- loidal metals, a catalytic action of metal hydrosols which produces effects similar to the ferments in the living organisms has been frequently suggested. Personally, I prefer to leave undecided whether such an expression as "catalytic action" has any real mean- ing in this connection or whether it is nothing but an empty word. We shall merely mention here the experiments with colloidal mer- cury, which has been chiefly used in syphilis and shows a specific action similar to that of other mercury preparations. Animal Experiments. In the case of silver hydrosol there exist many experiments of C. FoA and A. Aggazzotti.* They infected rabbits with staphyl- ococci and after an hour injected 30 cc. of a red silver hydrosol, repeating this several times. In this way they delayed the death of the animal from 1 to 3 days but recovery was not brought about. In infections with diplococci and typhoid (in dogs) the animals could be kept alive with injections of silver hydrosols. In the latter instance this was even possible when the silver-hydrosol injection was given in doses of 5 cc. intraperitoneally as late as 12 to 24 hours after the injection of the microorganisms. The same authors found that silver hydrosol has no effect on toxins in vitro, whereas it inhibits the toxicity if it is injected imme- diately after the toxin. From this they concluded that silver hydro- sol activates the oxidizing ferments of the body. Charrin, V. Henri and Monnier-Vinnard * speak very guard- edly concerning their therapeutic results, and characterize them as "very promising." Chirib and Monnier-Vinnard* experimented TOXICOLOGY AND PUARMACOLOdY 375 with pneumococci on white rats and mice. They ol>tained at times a retardation of the disease process and in individual instances they allege a cure by means of silver injections. Clinical Experiments. I shall pass over the majority of experiments which, because of their limited scope, are without significance and frequently contra- dictory, and shall only regard such results as are unimpeachable. To all appearances, only experiments performed with a stabilized silver hydrosol have practical value. The use of silver hydrosol, as collargol, in septicemia and pyemia is most frequent and best known. It is usually used as an intra- venous injection, at times as an ointment or an enema. If the numerous case histories^ are reviewed, two phenomena are prom- inent: the fall in temperature and the subjective improvement of the patient which follow several hours after the application is given. In contrast to this it is hardly possible to determine to what extent the disease process is influenced. The effect of silver hydrosol on pneumonia has been studied most thoroughly. G. Etienne* and J. Cavadias obtained good results; the rapid defer- vescence is also the most significant fact here. G. Izar*^ treated 28 cases of pneumonia with silver hydrosol and several with plati- num and iridium hydrosol; no difference was noted between the Ag, Pt and Ir. These thoroughly studied cases gave the followdng results: the course of the pneumonia process seems in general to have been favorably influenced though it was hardly possible to at- tribute this to a specific action upon the infectious process, but rather to the amehoration of the symptoms. As in the case of healthy individuals, in a pneumonia patient a rise of temperature, which reaches its maximum in about 4 hours, follows the injection and this is followed by a severe rigor, which is succeeded by profuse sweating and a rapid temperature fall, "critical in character, however, it cannot be termed a crisis." The subjective improvement of the patient is characteristic of the action of silver hydrosol. The brief period of oppression and anxiety which accompanies the rigor is succeeded when the temperature falls ]\v a feeling of well- being or euphoria. Cardiac and renal functioning are not affected, nor is there any action on the course of the pnemnonia process as far as may be determined from a change in the excretion of chlorids. 1 A very complete bibUography is given by Weissmann, Über Kollargol. Therapeut. Monatsh., Aug., 1905. Mentioned by Iscovesco, Presse Mddicale, May S, 1907. 376 COLLOIDS IN BIOLOGY AND MEDICINE G. IzAR reaches the conclusion that "the regular use of the injections shortens the course of the infection and seems to make it more favor- able." It was mentioned at the outset that the number of infectious diseases in which silver hydrosols as well as other metal hydrosols were employed is very great, and the opinions of the results very divergent; silver hydrosol, and at times platinum-hydrosol, have been employed in inflammatory rheumatism and erysipelas, in typhoid and para-typhoid, in appendicitis, furunculosis, phlegmons, anthrax, cerebrospinal meningitis, and scarlatina, dysentery and diphtheria, etc. As in the case of the diseases previously described, it affects the temperature curve though at times only temporarily, and there is frequently no influence on the patients subjectively. I have not as yet discovered in the literature any published cases of the use of silver hydrosols in tuberculosis; if they exist they are probably isolated instances. The reader may well get the impression that there do not exist for most diseases such thorough studies as G. Izar's *^ in pneumonia, and that on this account th ^ records of metal hydrosol therapy are incomplete. Mercury. Mercury has been used for centuries in syphilis. Since metallic mercury as such, as well as in the very finely emulsified form of blue ointment, is absorbed by the organism, there is no reason for expecting a very marked difference to result from the colloidal solution. The chemical firm of von Heyden manufacture a mercury hydrosol called Hyrgolum and a mercurous chlorid hydrosol called Calomelol, which may also be employed for inunctions. Sulphur. For some time a water-soluble sulphur hydrosol has been intro- duced into medicine and employed in skin diseases. Its action depends on the method of introduction since sulphur is reduced to the highly toxic hydrogen sulphid in the organism. The lethal dose for a rabbit weighing 1 kilo, according to L. Sabbatani, is 0.0066 gm. of colloidal sulphur intravenously (death is immediate) , whereas death occurs only after several hours when 0.25 gm. is introduced into the alimentary tract. The action also depends on the kind of animal; dogs are much less sensitive to sulphur than other experimental animals. The reduction and consequently the toxicity depends on the physi- cal condition; it is most intense in colloidal, less in amorphous, and least in crystalline sulphur. Moreover the toxicity is directly pre- TOXICOLOGY AND PHARMACOLOGY 377 portional to the dispersion. Joseph recommends sulphur hydrosol in diseases of the skin. Phosphorus, Arsenic, Antimony. Of all the complicated phenomena caused by these three sub- stances in different doses, there is only one which can be considered coUoid-chemically. Phosphorus, arsenic and antimony greatly influ- ence metabolism. Whereas arsenic and arsenic salts inhibit liver autolysis even in small doses, minimal doses of arsenic trisulphid hydrosol favor it. Small quantities of the latter preparation activate and larger ones inhibit the uric acid forming ferments in liver autol- ysis (M. AscoLi and G. Izar *). Phosphorus, arsenic and antimony inhibit oxidation processes. In minute doses this results in an increased constructive activity; its effect may be compared with sUght oxygen need, such as occurs at high altitudes. In larger doses the toxic action comes to the fore- ground. The metabolism does not reach its end product, weak car- bonic acid, but there are formed the intermediary stronger acids (lactic acid, glycuronic acid, etc.); the difficultly oxidizable fats are no longer normally attacked; there is a fatty degeneration of the glands (liver, kidneys), subcutaneous tissue and in the peritoneum and all the organs successively. [It is more probable that there is a change in the aggregation of the fat globules as the result of these poisons (breaking of emulsions). T. Brailsford Robertson has recently presented this view, and he refers to the fact that Gay and Southard observed the loading of the gastric epithelium with visible fat globules in animals which have experienced anaphylactic shock. Science N. S., Vol. XLV, No. 1170, p. 568 et seq. Tr.] It is upon this very retention of fat that the therapeutic employment of arsenic depends. It has been recognized a long time by the arsenic eaters of Steiermark and by breeders. [This may be due to the destruction of the protective action of an emulsostatic substance. Tr.] With toxic doses, when the formation of stronger acids instead of weak carbonic acid occurs, there must results an increased friction of the blood in the capillaries. As a matter of fact circulatory disturb- ances are among the most characteristic phenomena of phosphorus, arsenic, antimony and lead poisoning. " Generalized dropsy " (edema resulting from acid formation in the tissues, see p. 208 et seq.) is a symptom of chronic arsenic poisoning. We must also regard the "capillary paralysis" due to arsenic as caused by an increase of the viscosity of the blood at the inter- faces. It must be specially emphasized that these statements are only working hypotheses. 378 COLLOIDS IN BIOLOGY AND MEDICINE Salts. The neutral salts of alkalis may cause injuries^ to organs or organ groups by reversible changes in the condition of the organ colloids; strictly speaking, they are not poisons. We are unable to produce a poisoning, for instance by the oral ingestion of moderate doses of potassium salts, though this may be accomplished with intravenous injections; under such circumstances, disturbances of the heart muscle and the peripheral vessels are observed. It would be worth determining whether these phenomena are not to a great extent caused by changes in the viscosity of the blood. Hitherto, potassium salts have not been purposely employed therapeutically with this in view. H. Bechhold and J. Ziegler*'' attribute the favorable action of a vegetarian diet in gout to the generous supply of potassium salts which hinders the precipitation of urates. The biological action of neutral salts has been studied chiefly by biologists and physiologists. We owe to them valuable contri- butions concerning the inhibition of irritability (see p. 274 et seq.), the death of lower salt and fresh water organisms in changed media, and the inhibition of the development of the eggs of marine creatures. It follows from all these investigations that for the normal function- ing of the organisms, no matter whether animal or plant, high or low, a definite combination of electrolytes is necessary; upon this the normal state of swelling for the organ colloids depends. The cations are especially important. The monovalent cations (Na, K) are held in check by small quantities of divalent ones (Ca, Mg) . [See Clowes, p. 38. Tr.] Several examples may serve to explain this. For ani- mal organisms a given content of Na ions is necessary, which may at best be replaced by Li ions. K ipns are especially poisonous because they change the state of turgescence of the organ colloids. Pure sodium chlorid solution of physiological osmotic pressure behaves as a poison; this was shown by Jacques Loeb on the fertilized eggs of fundulus heteroclitus, a small sea anemone. He also showed that this poisonous action was arrested by the addition of a small amount of any salt containing polyvalent cations. Substances which were themselves very poisonous, such as barium, zinc, lead and ura- nium salts, under these circumstances detoxicate sodium chlorid, but copper and mercury salts and ferric ions showed no detoxicating action. K. G. Lillie *^ observed a similar antitoxic action of poly- valent cations in the poisoning of the larval forms of arenicola, a sea annelid. Its ciliary movement is stopped by pure Na and Li salts 1 These questions are treated in Chapter XVII. TOXICOLOGY AND PHARMACOLOGY 379 since the cilia dissolve. This injurious action is stopped by poly- valent cations. Interesting in this connection are the experiments of Wo. Ost- WALD *^ on the vitality of the sand flea (gammarus pulex) which lives in fresh water. It survives in sea water three or four days but in a mixture of four-fifths sea water and one-fifth distilled water, it lives almost as long as in fresh water. If each constituent of the sea water is successively removed, the toxicity of the remainder rises, that is the duration of life diminishes in the following order: NaCl + KCl + CaCl2 + MgS04 + MgClz NaCl + KCl + CaCl2 + MgSO« NaCl + KCl + CaCl2 NaCl + KCl NaCl. According to W. J. V. Osterhout what has been demonstrated for animals is equally true for plants (algae, grains, hverwort and moulds). The fresh water alga, vaucheria sessilis, is killed in -^^ NaCl solution but continues to grow if a trace of calcium chlorid ia added. According to Chas. B. Lipman the dry weight of ripe barley was increased if CaS04 was added to a culture containing sufficient sodium sulphate to be harmful. In this case as with cul- tures of bacteria, the antagonistic action of the cations play an important part. Though we employ physiological sodium chlorid solution in many experiments for the maintenance of isotonicity, it is merely a make- shift, and on this account there have recently been introduced solu- tions which, as well as being isotonic, have a composition similar to the blood (Ringer's and Adler's solution) and thus maintain its normal state of swelling. [More recently McClendon's. Tr.] All these solutions contain the divalent Ca ion. We have indi- cated on page 70 how we believe its detoxicating effect is brought about; it opposes the swelling due to monovalent ions (Na, K). And it is usually assumed that the "tanning" is hmited to the plasma pellicle. Though the cations are of major importance in "balanced" combinations of salts, the anions are not without significance (J. Loeb). As was mentioned previously, the toxic action of the neutral salts, is, in general, reversible. On this account the question arises, whether their action is due to a solution or an adsorption phenomenon by the organ colloids. Wo. Ostwald decided the question in favor of the latter view. In the adsorption equation (sec p. 21) in- 380 COLLOIDS IN BIOLOGY AND MEDICINE X stead of — (concentration of the salt in the dispersed phase) he placed - , in which t = length of life; - is accordingly the toxicity. 1 The equation becomes -r- = k. Wo. Ostwald experimented with the P sand flea mentioned (gammarus pulex) and with another small crus- tacean {daphnia magna). He placed a given number of them, e.g., twenty-five, in a definite quantity of water (100 cc.) of different salt concentration and every two minutes he observed how many had meanwhile died. It was evident that the zero point of the adsorp- tion curve must be placed to coincide with the normal salt con- tent of the organism, and that either a dilution or a concentration of the surrounding water is toxic. This must be expressed in the ad- sorption equation. Accordingly, the toxicity formula for neutral 1 salts, when their concentration is increased, is = k; in (c — n)~ V this case n is the quantity of salt normally adsorbed in the tissues. For the toxicity of subnormal salt solutions, the adsorption for- mula becomes - • C ~ = k. Wo. Ostwald ** calls the latter the t p "formula of leaching." Observed and calculated results agree quite well. A peculiar place is occupied by potassium iodid and iodin com- pounds. With all of them, the "iodin action" is the most im- portant; we may even assume that the iodin of nonelectrolytes finally becomes an iodin ion. The emaciation caused by its pro- longed internal use and the atrophy of certain glands are the most characteristic iodin effects upon higher animals. Prolonged use of iodin preparations, according to H. Meyer and R. Gottlieb,* among others, causes an excessive secretion from mucous mem- branes, which is an inflammatory reaction. Even though metab- olism experiments have not revealed any constant variations from the normal, it may be recalled that according to the experiments of H, Bechhold and J. Ziegler (see p. 54) potassium iodid facihtates the diffusion of a third substance through a jelly. All the phenomena mentioned above indicate a facilitation of metabolism. As was to be expected potassium iodid (according to E. Romberg) lowers the viscosity of the blood, and according to O. Müller and R. Inada* TOXICOLOGY AND PHARMACOLOGY 381 improves its circulation. The action of iodin in the functional dis- turbances of arteriosclerosis may be explained by this property since such disturbances may be attributed to a faulty blood supply to the organs. The analysis of the individual features of the process has not yet been completed. E. Bernoulli explains the action of bromin salts as a colloidal ac- tion. Bromids, which are given as sedatives, induce in both man and beast apathy and slumber as their most marked effect. It may be demonstrated that a portion of the chlorin in the body is displaced by bromin and that administration of NaCl induces recovery. E. Bernoulli has shown that the brain is more swollen in equimolecular solutions of NaBr than of NaCl. In addition he was able to restore rabbits poisoned with NaBr by injecting, instead of NaCl, other salts which inhibit swelling (sodium sulphate and nitrate). Thus it is highly probable that change in the function of the nerve cells in- duced by bromids may be attributed to a swelling. In the case of the alkaline earths there occur actual specific actions and we find transitions to irreversible conditions which are induced by the salts of heavy metals on albumin and lipoid colloids. For instance, barium has a very intense action on the heart and the vascular musculature. Of all the anions sulphocyanid inhibits pre- cipitation least, so that Wo. Pauli ** asserted, a priori, that a com- bination of sulphocyanid and barium would exert an especially severe effect. He maintained animals under the influence of a moderate sulphocyanid intoxication which, though the heart was strong and regular, stimulated the vagus and the vascular centers. In a moderate-sized dog 5 mg. of barium chlorid sufficed to cause an immediate stoppage of the heart. Calcium and strontium salts acted in a similar way, but much larger doses were required since with these there is much less specific affinity for heart muscle. C. Neuberg * and his pupils were able to prepare in methyl alco- hol colloidal solutions and jellies of compounds of calcium, strontium, barium and magnesium, which are insoluble in water, as for instance CaO, CaS04, CaCOs, the oxalate and phosphate of Ca, ]\IgHP04, BaCOs, etc. Since they are lipoid-soluble, it is possible they are of importance in the animal organism. C. Neuberg believes that pos- sibly they may develop in the cells in the presence of sugar, glj'cerin or even in the presence of ethyl alcohol in an aerobic respiration; in my opinion the presence of the body colloids should suffice to permit them to develop. The blood pressure elevating properties of barium salts may eventually be utilized in the form of colloid solutions in- asmuch as such solutions do not possess the undesirable by-effects of barium salts. 382 COLLOIDS IN BIOLOGY AND MEDICINE Aluminium is the bond between the earth alkalis and the heavy metals. It coagulates albumin in '' irregular series" and under certain conditions the albumin-aluminium precipitates are reversible. In this connection, thallium coagulates the protoplasm of aquatic plants (spirogyra, elodea, etc.), but they recover when replaced in their original medium (J. Szücs). The soluble salts of heavy metals form irreversible metal albumin precipitates with albumin which either flock out immediately or, depending on the concentration of the salt solution, persist in the colloidal condition. For this property of the salts of the heavy metals , besides the valence, the electrolytic solution pressure (see H. Bechhold *^) is determinative; colloid precipitation depends upon these two factors. The toxicity threshold of the various salts of the heavy metals has been arranged in series. Mathews * tested it on the motor nerves of frogs. Kahlenberg and True, as well as F. D. Heald, tested them on plant seedlings. I reproduce (from R. Höber) the series determined by Mathews for the inhibition of the develop- ment of the fertilized eggs of the sea anemone, fundulus heteroclitus. Salts. MnCl2 ZnCla CdCl2 FeCla CaCl2 NiCl2 Pb (CH3COO)2 CuCl2 HgCl2 AgNOg AuCls Solution pressure in volts. +0.798 +0.493 +0.143 +0.063 -0.045 -0.049 -0.129 -0.606 -1.027 -1.048 -1.356 Threshold of toxicity. 1/4 n 1/800 n 1/12,500 Ji 1/lOn l/12n 1/15 n 1/5,000« 1/15,000 n 1/50,000 n 1/90,000« 1/20,000« The exceptions which ZnCl2 and CdCl2 show (according to R. Höber) may depend in the first instance upon strong hydrolysis (acid reaction) and in the latter on the smaller amount of electro- lytic dissociation together with greater lipoid solubility. For the antagonistic action of ions of the heavy metals see pages 70 and 378. The intravenous injection of the salts of the heavy metals, which is associated with precipitation of protein, causes in suitable doses anaphylactic phenomena which may be explained by what has been said on page 210. The salts of the heavy metals in respect to their toxicity appear to me to have powerful specific influences. For instance, copper TOXICOLOGY AND PHARMACOLOGY 383 salts are powerful poisons to algae, infusoria and fungi. According to BoKORNY they are effective even in dilutions of 1 : 100,000,000. Vertebrates can stand them in relatively higher doses; but even among these there is considerable variation; cats, for instance, are said to be very sensitive to copper salts. Some of the heavy metal cations in spite of always precipitating al- bumin appear to be able to enter the circulation and to be definitely stopped only when they reach the filter membranes of the glands (liver, spleen, kidneys). On this account we frequently encounter kidney irritation from the toxic heavy metal cations (mercury, lead, etc.). Doubtless their solubilities in the hpoids are an important factor. The formation of irreversible albumin compounds kills the cell which is involved. [Hg, when absorbed to the extent of 4 mg. per kilo, slays relentlessly. It forms an irreversible compound, unaffected by antidotes or by washing with water as has been shown by Sansum. Tr.] On this account besides the acids and the alkalis, salts of the heavy metals, e.g., copper sulphate, silver nitrate and zinc chlorid, are used as caustics. Astringents act by causing a coagulation of the topmost layers of mucous membranes or inflamed surfaces. There- fore they include salts of the heavy metals, as silver nitrate, copper sul- phate and acetate, zinc sulphate and acetate and bismuth subnitrate. Besides these, ferric chlorid and the various aluminium salts (alumin- ium acetate, alum, etc.) of whose powerful flocculating action, resulting from the trivalence of Fe and Al we have already learned (see p. 84) ; the flocculating action in fact depends on the colloidal ferric hydroxid and aluminium hydroxid contained (see below). Similar results may be obtained with tannin, formaldehyd, and in short from all the hard- ening agents discussed in Chapter XXIII, provided their employment is not precluded by undesirable properties {e.g., picric acid and osmic acid) . Iron Salts and Iron Oxid Hydrosol. Recent researches have shown that only ionizable iron compounds have a pharmacologic action (upon the formation of red blood cor- puscles in chlorosis), but they show,^ on the contrary, that prepara- tions with iron firmly bound (hemoglobin preparations in particular) have no specific action. The numerous preparations in which iron is administered as a colloidal iron oxid {ferri oxidat. saccharatum solu- bile, liq. ferri oxid. dialys, and in some of the chalybeate mineral ^ It may be mentioned in contradiction to this, that colloidal Fe (OH) 3, ac- cording to M. AscoLi and G. Izar, favors the total autolysis of the liver as weU as its individual factors (see p. 369 et seq.) and that the ferments taking part in the formation of uric acid are activated by the addition of coUoidal ferric hydroxid; larger quantities, however, inhibit uric acid formation. 384 COLLOIDS IN BIOLOGY AND MEDICINE waters) are active only to the extent that they are dissolved in the hydrochloric acid of the gastric juice. I cannot form any idea as to the process of absorption since in the alkaline content of the small intestine where absorption occurs, the iron is thrown down again as a colloidal gel. Those colloidal iron preparations from which the iron ion slowly splits off (e.g., liquor Jerri albuminati, ferratin, etc.) are preferable since they exert a less injurious effect on stomach and intestine (indigestion and constipation) . After intravenous or subcu- taneous injection of iron salts colloidal ferric albuminate compounds are formed which may cause severe anaphylactic-Hke symptoms of poisoning (see p. 382) . When iron salts are taken by mouth this action does not occur, since the iron is arrested in the liver. The cathodal- migrating positive iron oxid hydrosol precipitates with the anodal- migrating blood colloids as an irreversible gel. This is the reason why ferric chlorid is so suitable for hemostasis. The greater part of the Fe in FeCls exists as iron oxid hydrosol as the result of hydrolytic cleavage. When blood coagulates, the excess of HCl is bound by the blood salts. R. BuNSEN, in his first scientific paper, showed that ''freshly pre- cipitated ferric hydroxid" is able to take up considerable quantities of arsenious acid and recommended it on this account as an antidote for arsenic poisoning. W. Biltz *2 showed that the distribution of arsenious acid between iron oxid hydrogel and water has the charac- teristic of an adsorption curve and not that of a chemical combina- tion. The protective action against arsenious acid depends moreover upon the method of preparing the ferric oxid hydrogel. Works on materia medica prescribe that it be freshly prepared. Perhaps, the inhibiting action which, according to L. Pincussohn,* ferric oxid hydrosol exerts on pepsin digestion depends upon adsorption. Although colloidal ferric hydroxid serves as the typical positive colloid H. W. Fischer *^ succeeded in preparing a negative ferric oxid hydrosol, as well. He did this by pouring ferric chlorid solution into sodium hydrate solution which contained glycerin as a protector. Glycerin and the excess of alkali were then removed by diffusion. Instead of glycerin other polyvalent alcohols, e.g., mannit, erythrit and cane sugar, may be employed. The object of his experiments was to obtain ferric oxid hydrosol which might be injected intrave- nously. Positive ferric oxid precipitates with the negative serum colloids; on this account the intravenous injection of positive ferric oxid is immediately fatal to animals, on account of embolism. A remarkable exception to this was found by C. Foa and A. Aggazzotti * in dogs; they are insensitive to positive ferric oxid; no explanation for this exists. Negative ferric oxid may be mixed mth serum in any proportion. It forms a deep ruby red solution which may at times TOXICOLOGY AND PHARMACOLOGY 385 take up much more than its own volume of oxygen. Since it has some other properties of hemoglobin H. W. Fischer calls this prepa- ration "synthetic active hemoglobin" {Effectsynthese des Hämo- globins). Properly prepared ferric oxid may be injected intravenously into rabbits; yet depending upon how it was prepared it proved to be more or less toxic even though no embolism could be discovered. Negative ferric oxid seems to store itself up in the glandular organs (liver, kidneys) just as do other hydrophobe, mostly negative col- loids. No change of charge occurs since it is only after HCl is added that a blue coloration occurs with potassium f errocyanid. Although positive ferric oxid hydrosol strongly adsorbs arsenious acid, its protective action is almost completely lost if such a mixture of the ferric oxid hydrosol and the adsorbed arsenious acid is injected subcutaneously. Negative ferric oxid hydrosol, under the same circumstances, exerts a very considerable protective action, but fails completely when such a mixture is injected intravenously. H. W. Fischer attributes this to the presence of hemoglobin which tears the arsenious acid from the ferric oxid hydrosol. Narcotics and Anesthetics. We class as narcotics such substances as temporarily suspend cerebral function, and the activity of the reflex centers. Narcosis is, therefore, a reversible process. According to the theory of Hans Meyer and E. Overton, nar- cosis is produced by such substances as dissolve especially easily in the lipoids of the plasma pellicle but are not entirely insoluble in the plasma.^ They determined the distribution coefficient between oil and water for a large number of substances and found that those substances in which the distribution coefficient (oil : w^ater) is high are good narcotics, e.g., chloroform, ether, acetone, chloral hydrate, urethan, etc. The coincidence is not only qualitative but it was possible by determining the "critical concentration" to show that it was quantitative. By "critical concentration" is meant the con- centration of a narcotic in water which just suffices to maintain the narcosis of an organism (animal or plant). With over 100 substances, a surprising parallehsm was shown to exist between "critical narcotic concentration" and coefficient of diffusion between oil and water, so that a causative connection between narcosis and fat solubihty seems obvious. ^ There exists a certain parallelism between the phj'^siological action of nar- cotics and their abiUty to depress the surface tension of water. Upon this is based J. Traube's * theory of narcosis. The depression of surface tension favors the penetration of the narcotic into the cell. 386 COLLOIDS IN BIOLOGY AND MEDICINE In recent years we have become acquainted with a number of facts which cannot be reconciled with the Meyer-Overton theory. For instance, S. J. Meltzer showed that magnesium salts possess power- ful narcotic properties. G. Mansfeld and Bosanyi then showed that during profound magnesium narcosis there was absolutely no change from the normal magnesium content of the brain. No increase in Mg was demonstrable either in the lipoid or the lipoid free brain substance. Furthermore, it developed, that the lipoid solubility of the narcotics was to a certain extent merely accidental which paral- leled other physico-chemical properties. According to J. Traube and J. Czapek diminution of surface tension parallels the narcotic properties. We must emphasize, however, that in Traube's experi- ments only the diminution of the surface tension to air was deter- mined, whereas in the organism we are concerned with surface tensions arising between two fluids or between a fluid and a gel phase. The observations of Battelli and Stern have less connec- tion with fat solubility; according to them there is a parallelism between the precipitation of certain proteins, the inhibition of oxida- tions in the tissues and the narcotizing activity of narcotics. War- burg and Wiesel showed that narcotics inhibit the ferment activity of the pressed juice of yeast as well as of the yeast cells. Without discussing the hypothetical basis of these processes we may conclude from them that lipoid solubility does not constitute the sole physico- chemical basis for narcosis. At present the tendency is to believe that the essential factor in narcosis is a modification of the plasma pelhcle which reversibly changes its normal permeability for electrolytes, so that it is an open question whether this membrane is pure protein (see p. 239 et seq., membrane) or a mixture of lipoid and protein (see also S. Loewe). An interesting support for this view was supplied by R. Höber and his pupil A. Joel when they measured the electric conductivity of blood corpuscles under the influence of narcotics. Although it is true that blood corpuscles are not nerve cells there are such similari- ties as justify us in applying to nerve cells, observations made on blood corpuscles. R,. Höber found that narcotics inhibited the exit of electrolytes when dilute, and increased it when concentrated. Nar- cotics when dilute produce quite the opposite effect they do when they are concentrated. This is analogous to the conductivity determinations of Osterhout on plant cells and the observations of Sv. Arrhenius and Bubanovic as well as J. Traube that small amounts of many hemolytic agents inhibit homolysis. Obviously, every substance which dissolves in fat is not a narcotic; it is such only if it can be again removed from the lipoid without TOXICOLOGY AND PHARMACOLOGY 387 leaving permanent changes. We thus arrive at the chief point in the problem. The Meyer-Overton theory explains the conditions under which a substance may act as a narcotic, but it does not show why it narcotizes; in other words, what the essence of narcosis is. Recent investigations, especially those of R, Höber, have shown that narcosis is brought about by a change in the state of swelling of the nerve colloids by which the changes which would otherwise be induced by the cell electrolytes upon stimulation are arrested. Ex- perimentally we consider an organ narcotized if its irritability is temporarily arrested or definitely changed. If we pass the im- pulse of an electric current through a muscle it contracts. If the ends of a muscle are attached to a galvanometer and we stimulate the muscle the needle of the galvanometer makes a short excursion; this is called the current of action. This is associated in no way with the muscular contraction, for we may produce an electric im- pulse in the nerve the same way and nerves do not contract. The excursion of the galvanometer needle is the only evidence that the nerve is stimulated. All these phenomena are temporarily arrested as soon as the organ is narcotized. If we now see that normal irritability is manifest as the result of an electrolytic process in which transitory changes in turgescence occur, and that the turgescence of nerves and muscle colloids are changed by salts, by which the irritability is consequently influenced, we shall not doubt that there is a connection between turgescence and irritability. When we find that the influence of salts upon the swelling capacity of cell colloids, especiaUy the hpoids, is placed in abeyance or suspended by narcotics, the mass of evidence is con- clusive. The connection between irritability and colloid turgor was dis- cussed in Chapters XVII and XXI; the following passages will show that narcotics arrest changes in turgescence. R. Höber *'^ has shown that the axis cylinders of nerve fibers swelled up in some portions under the influence of neutral salts and shrank in others, as is beautifully shown by staining with methylene blue. The phe- nomenon is reversible. Swelling under the influence of neutral salts does not occur when ethyl urethan narcosis is produced simultane- ously. Accordingly, in this case the narcosis may be demonstrated in the stained sections (see p. 336). A. R. Moore and H. E. Roaf* found that lipoid suspensions are precipitated by small quantities of chloroform, alcohol, ether, etc., instead of being dissolved by them. R. Goldschmidt and E. Pribram * found a similar action of chloral hydrate and urethan in lecithin suspensions. According to S. J. Meltzer magnesium salts produce narcosis if 388 COLLOIDS IN BIOLOGY AND MEDICINE subcutaneously or intravenously injected. I wish to call attention to the fact that according to 0. Purges and E. Neubauer * MgS04 and MgCl2 in — r- solution, unlike other electrolytes, have very nar- row precipitation limits for lecithin suspensions; with this fact their narcotic action possibly stands in some relation. Lower animals are also narcotized by magnesium salts. On this account it is used by zoologists to fix objects in their natural state, because Mg narcosis is not preceded by irritation. There is still not very much evidence that change in swelling is inhibited by narcotics; the evidence must be reinforced, especially by simple test-tube experiments on the rela- tive influence of salts and narcotics in changing the turgor of lipoids. We see here a promising field for experiment. It may be possible to combine this theory with that of Verworn's school. According to their view, the oxidation processes in the cell are arrested during narcosis, a hypothesis supported by numerous experiments. [A. R. C. Haas has recently shown that when Laminaria is exposed to anes- thetics (in sufficient concentration to produce any result) there is an increase in respiration, which may be followed by a decrease if the re- agent is sufficiently toxic. Science N. S., No. 1193, p. 46 e^seg. Tr.] In this connection it must be recalled, especially, that oxygen and carbonic acid are much more soluble in lipoids than in water and that narcotics diminish the absorption capacity of the cell fipoids for oxy- gen (G. Mansfeld*). It would be interesting to determine the extent to which this solubility is influenced by the turgor of the lipoids. Elsewhere I have already stated that the Meyer-Overton theory of narcosis demands a reversible distribution of the narcotic be- tween fipoids and plasma. Whether this distribution occurs as a Henry's distribution or as an adsorption is immaterial in principle (but not for the action!). According to a table of M. Nicloux * the distribution of chloroform seems to me to approach that of adsorp- tion. After the termination of narcosis, the blood of a dog con- tained the following content of chloroform (in per cent). Chloroform Content in Per Cent. After. First experiment. Second experiment. 0 minutes 0.054 0.0255 0.0205 0.018 0.0135 0.0595 15 minutes 30 minutes 6.023 1 hour 0.018 3 hours 0.0075 7 hours 0.0015 TOXICOLOGY AND PHARMACOLOGY 389 The action of narcotics on the permeabihty of electr-olytes is reversible according to R. Höber, and may be reversed by washing them out provided the amount of added narcotic is not too great. R. Höber is of the opinion that narcosis is characterized by a change in the plasma pellicle in which the increase of permeability to normal stimuli is inhibited. A physico-chemical study by S. Loewe actually showed that chloroform was adsorbed by the white matter of the brain and that sulphonal, trional and tetronal were adsorbed by lipoids. We see from the table that, at first, chloroform disappears very rapidly but that the final portions are tenaciously held. A similar table for ether reveals an approximately proportional disappearance of ether from the blood in given units of time, which would approxi- mately answer the demands of Henry's law. The slower recovery from chloroform than from ether narcosis is thus explained. It is evident from what has been previously said that narcosis merely represents a given segment of the curve which different con- centrations of the narcotic cause in the turgor of the cell lipoids. The conmiencement of the curve with low narcotic concentration indicates the condition of irritability before narcosis, the terminal limb with high narcotic concentration means death. What has been said here of benumbing the entire body mutatis mutandis, appHes, for the individual organs, in the case of local anes- thesia. Local anesthesia may be produced by all sorts of substances — by very dilute caustics (acids, phenol), by distilled water, by aniso- tonic salt solutions, in short, by all substances which change the turgor of the cell hpoids. Practically most of them are useless be- cause the first portion of the curve, the state of irritation which is expressed by pain in subcutaneous injections, is too prolonged; in the case of others, because the segment which signifies local anesthesia and which hes between the ''irritation limb" and that of permanent damage is too short; still in others, because an irreversible change in the cell colloids may occur even with the smallest doses, or other cell colloids suffer too much in sympathy. Practically only such anes- thetics are utihzable as produce only a reversible change in the turgor of the nerve lipoids, as is exempHfied by cocain, novocain and anesthesin. It is not difficult to range the other methods of anesthesia, such as cold and the production of anemia in this scheme, but experimental confirmation is still lacking. Colloid research also offers an explanation of certain by-effects of narcotics. [Evarts A. Graham has shown that the toxic action of many anesthetics is due in part to mineral acids formed by their 390 COLLOIDS IN BIOLOGY AND MEDICINE decomposition. He believes that delayed chloroform poisoning results entirely from the destructive action of HCl formed in the tissues and he attributes the protective action of glycogen to the fact that the glucose resulting therefrom inhibits the diffusion of HCl into gels. The toxic action of anesthetics has been shown by J. A. Nef to be due to an unsaturated carbon atom. The effect of such atoms has not yet been discussed colloid-chemically. (Jour. Amer, Med. Assoc, Vol. LXIX, No. 20, p. 1066 et seq., quoted by Graham, loc. cit.) , . Bürge attributes the anesthetic action of anesthetics to the de- crease in oxidation processes produced by the destruction of catalase. The specific action on the nervous system is due to the greater solu- bility of the lipoids of nervous tissue facilitating the entrance of the narcotic into the nerve cell. Science N. S., Vol. XLVI, No. 1199, p. 618 et seq. Tr.] With large doses of morphine, chloroform and ether we observe more or less intense phenomena of irritation, especially in the kidneys, before the general circulation is much disturbed; album- inuria and hematuria may thus occur. Martin H. Fischer * (see p. 333) explains this by the disturbance in the oxidation processes of the body which suffers from such substances and by the fact that as the result of the accumulation of CO2 and ultimately of other acids, a fixation of water occurs in the body so that no excess of water remains for excretion by the kidneys. Besides the anuria, we may thus explain the thirst which such patients frequently show. Secre- tion of urine occurs again and the thirst disappears when the effect of the narcotic wears off, even though the patient takes no water. Small doses of ether, alcohol, etc., cause the reverse phenomenon, since by increasing the activity of the heart they bring on an im- provement in the supply of oxygen. By this means not only a stronger flow of blood is suppHed to the kidneys but the "free" filterable water in the blood is increased, provided the oxidation processes are still uninjured. Colloid research seems to me to have raised new questions regarding investigations of the effects from the prolonged use of alcohol. Though the larger part of the alcohol ingested is seized by the lipoids, we cannot neglect the effect upon the albuminous colloids. At present we can only assume that it causes a diminution of swelling. The extent of the relationship between the degenerative changes of the cells, arteriosclerosis, etc., and of this action of alcohol remains for future investigations to determine. [W. Burridge has shown that alcohol increases the utilization of calcium by certain cells. Tr.] TOXICOLOGY AND PHARMACOLOGY 391 Disinfection. By disinfection we understand the killing or rendering harmless of dangerous germs outside of or within the body. Substances which destroy germs living on foods, without being very harmful to higher organisms, are called preservatives. For simplicity we shall first consider external disinfection by chemi- cal means. In the process of disinfection a distribution of the dis- infectant between the organism and the medium first takes place. This distribution may occur either in the manner of chemical combina- tion, adsorption or in accordance with Henry's law. In the two former cases it is conceivable that even traces of the poison are active, whereas this would be possible in Henry's distribution only if the substance is very much more soluble in the bacillus than in its medium. It follows from the ease with which they are stained that surface attraction is of great importance in the case of bacteria. And in fact staining and disinfection are distinguished only by the fact that in the latter instance the absorbed substance exerts a par- ticular poisonous action on the microorganism. If for the present we consider a microorganism only as a small particle without special chemical properties and add to such a hypo-, thetical emulsion of bacteria, a dissolved substance, this substance would by reason of the mere surface attraction have a tendency to concentrate on the surface of the bacteria to a greater or less ex- tent, depending upon the nature of the dissolved substance, i.e., the more strongly the given substance diminishes the surface tension of the water, the greater is the concentration at the surface.^ Most bacteria act like a suspension which has been protected by a protective colloid, before being flocculated by neutral salts; they are so changed by boiling or by agglutination that they change from hydrophile to hydrophobe suspensions, which cannot be differentiated physically from kaolin suspensions or the like. The electric charge is that of an inorganic suspension, i.e., negative; it is discharged by agglutinin. All these questions are taken up in detail in the chapter on "Immunity Reactions." As the dispersed phase, microorganisms are strongly adsorbed by substances with great surface development. (See Fig. 52.) Be- cause of this adsorption, they are readily held back in fine-pored filters such as Chamberland candles (unglazed porcelain), Berkefeld filters (Kieselguhr), asbestos, wadding or carbon filters. 1 This conception was originally developed and established by H. Bechhold in the "International Congress for Apphed Chemistry," London (May to June, 1909) (see KoUoid Zeitschr., 5, 22, 1909). 392 COLLOIDS IN BIOLOGY AND MEDICINE Besides the microorganisms directly visible in the microscope, there are others so small as to be microscopically invisible, and only recognizable by their pathogenic effects. They are, therefore, called ultravisible. Among these are about forty pathogenic germs, among others, smallpox, rabies, measles, scarlet fever and the mosaic disease DISINFECTING ACTION OF HALOGEN NAPHTHOLS. 400,00(V 350.000 Lethal action on 10,000 Number of halogen atoms. Co// — Sfaphy/ococcus -■> — — — Diphf-heria — X — ^x— tX—— Sfrep/ococcus — Pamtyphus Fig. 52. (See p. 402.) of tobacco. The name ultravisible is not a happy one, since recently by dark-field illumination there have been recognized, in the case of many infections, minute organisms, which we are justified in believing to be the cause of the diseases. The ultravisible viruses are not held back by ordinary bacteria filters; recently they have been called filterable microorganisms. TOXICOLOGY AND PHARMACOLOGY 393 The study of these forms of Ufe is difficult because of the lack of technical methods for their investigation. Besides dark-field illumi- nation, colloid research has provided two methods which have already led to important advances: these are ultrafiltration and adsorption. By means of the Chamberland filter the solution of virus may be freed from visible bacteria. In order to concentrate the filterable germs and make quantitative tests with them, they may be con- centrated on an ultrafilter, as was done by Betegh with hog cholera virus, Prowazek and Giemsa with variola; or they may be adsorbed on charcoal or clay (as did Gins with smallpox). I beheve the colloid investigation of filterable microorganisms will yield valuable results, since they form a transition group to true colloids. A beginning has already been made. Thus Andriewsky has shown by ultrafiltration that the virus of chicken cholera is smaller than the hemoglobin molecule. It has been repeatedly observed that the development of micro- organisms is facihtated by the presence of suspensions or hydrogels. Thus Krzemieniewski found that a pure culture of nitrifying bac- teria grew more luxuriantly and bound more nitrogen if earth or humus was added to the culture medium and Kasserer found a similar effect from the addition of colloidal siUcates and phosphates of iron and aluminum. According to Ross van Lennep pieces of kidney, meat, cellulose, etc., improve the growth of aneorobic bacteria, yeast and B. coU. We thus see that these microorganisms on purely physical grounds find much more favorable conditions for growth in their natural habitat than in artificial media. In some instances it was possible to determine the reason for this phenomenon. Thus Sohngen * and also Ross van Lennep showed that charcoal and some other sohds favor the dissipation of carbonic acid which inhibits the growth of yeast. In other instances the suspensions or colloids adsorb oxygen for aerobic bacteria, nitrogen for nitrifjöng bacteria, or other nutritive ingredients which are then available for growth at the surfaces of the respective substances (Hterature given by Sohngen *) .^ H. Freundlich *^ mentions the following substances which show sHght adsorptive affinity: salts (especially of the baser metals), highly dissociated substances (such as strong acids and bases), aggregations of OH groups (sugar) and the sulpho group. As a ' Though it is shown on page 396 that the distribution of phenol between the bodies of the bacteria and their environment occurs as it would in two sol- vents, it does not by any means contradict what has been said here, since a dia- infectant action does not result from adsorption. Disinfection occurs when the disinfectant penetrates the microorganism; the portion which has penetrated may very well comply with Henry's law (distribution). 394: COLLOIDS IN BIOLOGY AND MEDICINE matter of fact but few disinfectants are furnished by the inorganic acids and bases and by the salts of the baser metals. Of course we do not include such concentrations of the acids and bases as produce» a direct destruction of the organized substance. As a matter of fact, substances containing the phenyl group are our most useful dis- infectants, such as carbolic acid, cresol, naphthol, anilin water, etc. H. Bechhold and P. Ehrlich * by combining phenyl groups (derivatives of dioxj^diphenylmethan and o-diphenol) obtained sub- stances of hitherto unequalled disinfectant action (with the exception of sublimate, etc.) and even this action was greatly increased by .the introduction of halogens. The work of H. Bechhold,*^ which introduced into practice the halogen derivatives of naphthol and dicresol, disinfectants of great activity, establishes the breadth of this assumption. A dilute solution of alkalis or acids is the normal environment for the majority of microorganisms. Although the majority of micro- organisms prefer a more alkaline nutriment corresponding to the dearth of H ions in the animal organism, there are other bacteria and moulds, for instance, lactic acid bacteria, which require or prefer an acid medium, e.g., the moulds which grow on acid fruit. From this it follows that when acids or alkalis injuriously affect a micro- organism, the specific vital conditions of the microorganism in ques- tion have been unfavorably disturbed and accordingly it is impos- sible to speak of a general injurious action of H or OH ions. Many salts of the heavy metals (e.g., silver, mercury and copper salts) are disinfectants. Their strong adsorptive power, in which sublimate excels all others, was demonstrated by P. Morawitz.* Adsorptive capacity is only a condition preliminary to the exercise of specific toxic action. It is generally accepted in the case of salts of the heavy metals that this toxic action depends on the forma- tion of albuminates. I am at present engaged in the explanation of these phenomena and I am already in a position to state that adsorp- tion is by no means the most important factor. Finally, there are among the inorganic salts, substances with specific activity, e.g., the fluorids, thallium carbonate, sulphurous acid salts, boric acid, etc. We know of no disinfectants among the sugars or their related substances (e.g., glycerin). P. Ehrlich and H. Bechhold * as well as H. Bechhold *^ have shown in the case of a large number of aromatic compounds that the introduction of sulpho groups into a disinfectant considerably diminishes its activity. Adsorption in water according to H. Freundlich *^ is favorably influ- enced by the phenyl group and the halogens. This author mentions as an example chlorbenzoic acid (\ = 154), benzoic acid (X = 140). TOXICOLOGY AND PHARMACOLOGY 395 Microorganisms. Microorganisms occur more or less densely in their media as millions of minute dots, rods or threads. They constitute a dis- persed phase and as such obey the physical laws to which all suspend sions are subject. Collectively they possess an enormous develop- ment of surface; and, consequently, surface attraction especially influences those substances that are dissolved by them (in other words, more as the substance diminishes the surface tension of water) . Should our assumption that adsorption plays an essential part in disinfection be correct, then the same substance will be a much better disinfectant in aqueous solution than when dissolved in alcohol or in acetone.^ This assumption is sustained by such investigations as have been undertaken. According to Robert Koch, anthrax spares were not destroyed by the appKcation for 100 days of 5 per cent car- bolic acid in oil nor by 5 per cent carbolic acid in alcohol for 70 days, whereas they were destroyed after 48 hours' exposure to 5 per cent aqueous solution of carbohc acid. Anthrax baccilli were of undimin- ished virulence after 2 days' treatment with 5 per cent carbolic acid in oil, whereas 1 per cent aqueous solution killed them in 2 minutes. Moreover, according to Reichel,* the distribution of the phenol between albumin and the oil (as compared with water) is in favor of the oil. According to the researches of Paul and Krönig, as well as those of Sheurlen and K. Spiro, phenol acts in disinfecting as a molecule and not as an ion. Sodium carbolate which is strongly dissociated has a much weaker action than phenol. Phenol is less dissociated in alcohol than in water, so that if it were merely a question of dissociation, phenol should be a better disinfectant in alcoholic than in aqueous solutions. As is shown by the following data taken from Paul and Krönig's paper, the facts are quite the reverse. Anthrax spores were treated with the disinfectant, according to the marble method, and then sown on agar; the resulting colonies were counted. Number of colonies. 4 per cent carbolic acid in water 1505 4 per cent carbolic acid in alcohol oo We thus see that in disinfection adsorbability from water is more important than solubility. 1 In disinfecting the hands and skin, alcohol and alcoholic solutions and even acetone are almost exclusively used, though entirely different factors are of im- portance in determining their use (better capacity to wet the fatty epidermis, the shrinking action of alcohol and deeper penetration into the capillary spaces of the skin (Bechhold)). 396 COLLOIDS IN BIOLOGY AND MEDICINE Cresol is less soluble than phenol and is a stronger disinfectant than the latter. Its solubility in water is so limited that it must be dis- solved with the aid of soaps and similar substances. These are not true solutions; they are manifest emulsions in the dark field (Frei and Margadant), It is still an open question whether the effect on bacteria is exerted by an envelopment by the individual cresol soap droplets, thus forming about them a highly concentrated disinfectant film. Another possibihty is that the bacteria withdraw dissolved cresol from their environment, and that cresol diffuses from the droplets to an equal extent into the water. A group of disinfectants are active even in a dilution in which the substance is no longer chemically demonstrable. According to R. Koch, interference with the growth of anthrax bacilli is caused by sublimate even in a dilution of 1 : 600,000. According to H. Bech- HOLD and P. Ehrlich* tetrachlor-o-diphenol interferes with the growth of diphtheria bacilli in a dilution of 1 : 400,000 to 1 : 640,000. According to H. Bechhold,*^ tribrom-naphthol inhibits the growth of staphylococci in a dilution of 1 : 250,000. We can understand the effect of such traces of substances if we consider the course of the adsorption curve (see p. 20) in which the distribution between ad- sorbent and solvent occurs in such a way that the dissolved sub- stance is practically completely adsorbed in the weaker concentra- tion, whereas in higher concentrations the distribution approaches that required by Henry's law (between two solvents). The objection may be raised that the same conditions are fulfilled in a purely chemical combination, to which we may reply that in many instances such a chemical combination must be considered to occur. In favor of adsorption, there are two distinct phenomena, inhibition and death. By choosing a suitable disinfectant in sufficient concen- tration and exposure, microorganisms may be completely killed; that is, they cannot under any circumstances be brought back to life. In other cases, it is only necessary to remove the disinfectant, to dilute it or to transfer the germs to another environment, for the germs to start multiplying again; such action is called inhibition. In such a case, we must assume that the reaction between microorganism and dis- infectant is reversible. In killing, the process may be irreversible.^ 1 I can readily imagine that death may occur in a reversible process if the action of the disinfectant persists for a sufficient length of time to nullify other vital processes. To give a very crude comparison, if a man is drowned, the water cannot be regarded as a poison though it depresses necessary vital proc- esses. A man who cannot be resuscitated after a submersion lasting 5 minutes has fixed no more water in his body than one who has been resuscitated after 2 minutes' submersion. TOXICOLOGY AND PHARMACOLOGY 397 If the disinfectant were a firm combination with the microorganism it would be difficult to explain how the germ could multiply again when removed from the disinfectant solution. This is readily under- stood if we assume that the union between microorganism and dis- infectant is an adsorption. In that case the disinfectant will pass into the absolutely indifferent solvent so that the microorganism having become free again (from the disinfectant) is in a condition to continue its development. A few examples will explain the foregoing. R. Koch performed certain experiments in the following way: he dried germs on silk threads and subjected them for a given time to a disinfectant solu- tion; after this he placed them in nutrient bouillon or in gelatin; if the germs developed, he considered that the disinfectant was active; if they did not, that it was inactive. In this way R. Koch subjected anthrax spores for two days to 5 per cent carbohc acid and found that afterwards they did not develop in gelatin. B. Riedel, in the Imperial Health Office, found that, even after 14 days of im- mersion in 5 per cent carbolic acid, the germination of anthrax spores was not inhibited if the silk threads were first washed with water and then placed in fluid gelatin; the gelatin and silk threads were thoroughly mixed by prolonged agitation of the test tube. According to R. Koch, a single immersion of anthrax spores in 1 : 5000 sublimate solution suffices to destroy them. J. Geppekt * found that the same concentration acting four seconds longer, on one trial, produced their death and on another did not. Among eountless experiments on this point we shall mention those of Eisen- BERG and Okolska because of the method they employed. They mixed uniform quantities of disinfectant and bacteria, some- times adding the entire quantity of bacteria at once, and sometimes in fractions. If the phenomenon is reversible, the results in both cases should be the same; if it is irreversible there should be a point in the fractioning experiment when the disinfection should prove less satisfactory. As was to be expected from other considerations, the action of phenol proved to be reversible and that of KMn04 and HgCU to be partly irreversible (in these instances the time of action was an important factor). Numerous experiments have been performed in an attempt to test quantitatively the views given here; the results actually satisfy the hypothesis in some instances. An exact agreement between obser- vation and calculation is not to be expected because in disinfection adsorption is not the only factor, though it is chiefly accountable for the action of the disinfectant on the microorganism (lipoid solu- bility, modification of protoplasm, etc.). 398 COLLOIDS IN BIOLOGY AND MEDICINE The question of adsorption may be solved m one of two ways which I shall call respectively the chemical and the hiological methods. The chemical method regards the microorganisms as a lifeless sus- pension. Suspensions are shaken with various laiown dilutions of the disinfectant, and after the suspension is removed the amount of disinfectant remaining in the fluid is chemically determined. From this we learn how much has been absorbed by the microorganism in the various dilutions. It is the same method that is usually emplo^^ed in chemical adsorption experiments. It may be criticized because it determines the amount of disinfectant absorbed by the microorganism but not the result of the adsorptive action, the disinfection. From a concentrated solution much more disinfectant is removed than is necessary for killing or inhibition. R. 0. Herzog and Betzel * employed the chemical method, with yeast as the microorganism. They obtained an adsorption curve for chloroform and silver nitrate and a chemical combination for formaldehyd. The results are interesting inasmuch as chloroform obviously acts by reason of its lipoid solubility; I question whether the precipitation of albumin by silver nitrate is the only factor which deterixdnes its disinfectant action. The result for formaldehyd is especially surprising; its powerful inhibitive action on development is well known, however its lethal action was discovered to be much weaker. We shall await with great interest the further prosecu- tion of Herzog's experiments which promise an explanation of some of the questions proposed. The results with -phenol are quite compli- cated. According to Reichel,* in an aqueous solution of phenol there is a distribution in accordance with Henry's law, i.e., as if it were distributed between two solvents. This was demonstrated by Reichel * in the distribution of phenol between water and oil, albu- min, Cholesterin and the bodies of bacteria. This explains why phenol is active only in relatively high concentration. Increasing NaCl con- tent shifts the relative distribution in the direction of the nonaqueous phase. According to Reichel the disinfectant action depends on the fact that phenol causes a shrinking of the albumin phase; this is strengthened by the NaCl. In this way, the views developed by K. Spiro and J. Bruns * are revived in modified form. R. O. Herzog and Betzel obtained an adsorption curve on treat- ing yeast with a phenol solution weaker than one per cent. These contradictory results may probably be explained by the primary a(isorption of the phenol at the surface of the bacterial cell which then in some way absorbs it until the body of the bacterium is filled. This I infer from the experiments of E. Küster and Rothatjb TOXICOLOGY AND PHARMACOLOGY 399 who show that upon the death of the bacteria a part of the phenol is Kberated. The biological method regards the rapidity of death (measured by the number of surviving bacteria) in known concentrations of the disinfectant and during a known time for action. In this case, the changes in concentration by means of the adsorbing microorganism are not considered, as in the chemical method, but only the damage to the microorganism. The method assumes that "the rate at which the solution of a substance acts as a disinfectant is proportional to the amount adsorbed from this solution" (Morawitz*). This method also is open to the objection that microorganisms are noi a single mass with uniform vitality but a mixture in various stages of growth and with varying resistance; so that it is possible that the curves obtained do not represent the course of an adsorption in various concentrations, but express the resistance at various stages of growth. These criticisms are offered to show the difficulties encountered in an experimental test. We may count in this group, also, the experiments in which an insight into the mechanism of disinfection may be obtained, by varying the number of bacteria with known changes in the concen- tration of the disinfectant acting for a constant time (Eisenberg, Okolska) . As a result of biological methods, Paul, Birstein and Reuss * came to the conclusion that the death of dried adherent staphylococci in oxygen or in mixtures of oxygen and nitrogen is due to the adsorption of oxygen by the cocci. P. MoRAWiTz^ (loc. cit.) found a good agreement between the figures obtained by Krönig and Paul, upon kilhng anthrax spores with sublimate and the formula for adsorption. Accordingly, we learn from the quantitative tests that the dis- tribution of a disinfectant between microorganism and solution may possess the formula of a chemical combination (formaldehyd) of adsorption (chloroform, silver nitrate) and of distribution in solvents in accordance with Henry's law (phenol). In the following pages we shall see that transitions between these different kinds of distri- bution occur. It would certainly be an error to regard distribution as the essential factor in disinfection. As a result of adsorption the germ is sur- rounded by a highly concentrated film of disinfectant whose action ^ This calculation is referred to in the communication of H. Freundlich which is mentioned in the paper of H. Bechhold * on Disinfection and Colloid Chem- istry, page 23. 400 COLLOIDS IN BIOLOGY AND MEDICINE destroys it much sooner than would be expected from the extremely- dilute solutions employed; the germ retains the disinfectant in other media or in an infected organism and only subsequently succumbs to the damage the disinfectant inflicts. We must seek the essential activity of the disinfectant in a modification of the living substance with which the disinfectant combines or changes so that its vital function is suspended. I know of no experimental investigations which show what part of the disinfectant is combined (fixed) and what part is adsorbed, though such studies are very desirable as they would afford us clearer in- sight into the nature of disinfection, and they would also be of great practical significance. For the present we must be satisfied with analogies which without question can be applied correctly to the principle of disinfection. Chemically, the microorganisms have so much similarity to textile fibers, especially with wool and silk (to mention only the great similarity in staining), that we may properly employ in argument the results of W. Schellens. He shook 1 gm. of fiber with 50 cc. of a subhmate solution containing 1 per cent Hg and found : Hg fixed. Hg ad- sorbed.' Hg re- moved from the solution. From sublimate (containing 1% Hg) : Fruit hairs of eriodendron Per cent. 1.20 1.69 1.9 5.89 indeterminable trace u li 0.5 6.5 5.2 9.8 12.3 Per cent. 3.91 3.08 4.14 12.36 3.14 3.0 3.5 4.55 8 6.8 7.7 8.2 Per cent. 5.11 Jute 4.77 Silk 6.04 Wool 18.25 From mercuric cyanid (containing 1% Hg): Fruit hairs of eriodendron 3.14 Jute 3.0 Silk 3.5 Wool 5.05 From mercuric acetate (containing 1% Hg): Fruit hairs of eriodendron 14.5 Jute 12.0 Silk 17.5 Wool 20.5 1 The adsorption figures were calculated by me from the figures of Schellens. These figures are interesting from various points of view. We see that in the case of sublimate, of 3 parts of Hg, approximately 2 parts are adsorbed and only one is fixed. Mercuric cyanid is the one substance which is only adsorbed and suffers practically no fixation ; though it is true it powerfully inhibits development, it has but a TOXICOLOGY AND PHARMACOLOGY 401 weak destructive action. According to K. Spiro and J. Bruns,* as well as Paul and Krönig, the figures show that mercuric cyanid is far inferior to sublimate as a disinfectant. We see from the table, moreover, in the case of mercuric acetate, which is more strongly fixed and more strongly adsorbed than HgCl2, that fixation and adsorption are not in themselves alone sufficient for strong disinfection; the disinfectant must be offered in a suitable form. Mercuric actetate is less ionized than HgCl2, and since, according to Paul and Krönig, as well as Scheurlen and Spiro, the Hg ion is responsible for the disinfectant action, mercuric acetate is weaker than sublimate. An especially convincing proof of the specific chemical action of the disinfectant on the hving substance seems to me to be that there is a difference in the resistance of various groups of bacteria to disinfect- ants. Whereas anthrax spores, tubercle bacilli, etc., show an enor- mous resistance, cholera vibrios, gonococci and streptococci succumb to even sHght chemical attacks. The other groups of bacteria are ranged between these two extremes — typhoid, B. coli, staphylo- cocci, diptheria bacilli, etc. Were merely the strength of adsorption responsible for the disin- fectant action, we could readily understand that substances of differ- ent disinfectant power would exist; we would understand for instance that cresol has a stronger action than naphthol, but in that case cresol would always possess a stronger action than naphthol, both on B. coli and on typhoid bacilli, as well as on streptococci. If we found, how- ever, that lysol was more active against one microorganism and that /3-naphthol was more active against others, we could attribute the action to general physical properties among which we might include adsorption, but we would then have to ascribe it to the difference in behavior caused by specific inherent chemical differences in the bac- teria affected. This might be either a variation in the solubility of the bacterial pellicle or a variation in the grouping of the atoms in the body of the bacteria so as to manifest a greater or less affinity to the disinfectant; in either case the important factor is the chemical difference in the microorganism. Such cases actually exist as has been demonstrated by H. Bechhold.*^ He showed that the minimal lethal dose in 24 hours is: For lysol (the cresol content being com- pared) For /3-naphthol Diphtheria bacilli. 1 : 20,000 1 : 10,000 B. coli. 1 :800 1 : 8000 402 COLLOIDS IN BIOLOGY AND MEDICINE In accordance with this, lysol acts twice as powerfully against diphtheria bacilli as does (8-naphthol, whereas it has only one-tenth the effect of the latter on B. coli. He showed further that a mixture of tri- and tetrabrom-/3-naphthol in one per cent solution killed staphy- lococci in from two to three minutes, whereas lysol dilutions containing one per cent cresol took more than ten minutes to do so. Conversely, a 5 per cent lysol solution containing 2.5 per cent cresol is lethal for tubercle bacilli within four and a half hours, whereas a solution of tri- and tetrabrom-/3-naphthol of corresponding strength had no effect even at the end of twenty-four hours. We see, therefore, that tri- and tetrabrom-/3-naphthol surpass cresol in its action upon streptococci, while upon tubercle bacilli the cresol acts more power- fully. H. Bechhold *^ examined naphthols containing 1, 2, 3 or more bromin or chlorin atoms with reference to their effect on various bacteria. He found, that with the admission of the halogens the effect upon various bacteria sometimes increased, that at times it decreased, and that certain optima could be obtained (see Fig. 52 on page 392). Thus the maximum disinfectant action against staphy- lococci is obtained with tri- and tetrabrom-;8-naphthoV while for B. paratyphoid it is obtained with dibrom-jS-naphthol and so on. Eisen- BERG has recently determined partly specific activities for a large number of coal-tar dyes. It follows from this that to test an antiseptic on only one kind of bacteria is an absolutely inadequate method for testing disinfectants; it is necessary to subject a number of different types of bacteria to investigation. The presence of a third substance is a factor in the action of a disinfectant that cannot be neglected. We have already called at- tention on page 383 to the influence of the solvent. To Paul and Krönig, as well as to Scheurlen and Spiro, belongs the credit of having made clear the significance of electrolytic dissociation for disinfectant action. Dissociation may be increased or diminished by adding certain substances to the disinfectants. The ionization of HgCl2 is decreased by the addition of NaCl, and since it is the Hg ion which is of importance in disinfection, the addition of common salt diminishes the disinfectant action of sublimate. On the other hand, the disinfectant action of carbohc acid, cresol and the other phenols is decidedly increased by common salt. Since NaCl can have no effect on the electrolytic dissociation of phenols, we must seek some other explanation. Again, the nearest comparison must 1 Tribrom-;8-naphthol is sold under the trade name " providoform " by the Pro- vidogesellschaft (Berlin) and has proven useful in connection with the pus cocci and diphtheria bacilli. TOXICOLOGY AND PHARMACOLOGY 403 be drawn from the process of dyeing: common salt or sodium sul- phate is frequently added in dyeing cotton in order to get a more rapid and complete utilization of the bath. The simplest explana- tion of this is that there has been diminution in the solubility of the dye by means of the added salt (i.e., the dye is made more colloidal) and as a result of this a stimulation of adsorption occurs. This idea guided Spiro and Bruns * in their experiments. They found that salts and other substances which did not ''salt out" phenol from aqueous solutions, such as sodium benzoate, urea, glycerin, etc., had no effect in strengthening the disinfectant action of phenol. Pyro- catechin may be precipitated by ammonium sulphate but not by common salt; the former increases the disinfectant action of pyro- catechin, while the latter does not. It is also interesting that according to Paul and Krönig equimolecular quantities of salts added to a 4 per cent carbolic solution increases its action in the fol- lowing order: NaCl > KCl > NaBr > Nal > NaNOs > CsHgONa. According to Spiro and Bruns,* the same order obtains for the precipitating action of these salts on phenol; however, the sulphates exert a much more powerful effect. The close relationship of this series of salts to albumin precipitation and to many other biological processes is quite obvious (see pp. 81 and 272). Frei and Marga- DANT have determined similar relations between both the increased activity of cresol soap solutions by salts of the hght and of the heavy metals as well as the decreased surface tension induced by such salts. We may imagine that there is yet another possible way for salts or other substances to exert an influence by their mere presence. H. Bechhold and Ziegler*- showed that the permeability of jellies was influenced by certain substances, and from this we may assume that the permeability of the bacterial plasma pellicle for a disin- fectant may be changed by the presence of a third substance. This assumption is reinforced by experiments of Eisenberg and Okolska which showed that alcohol, alkalis, urea and some other substances, which increase the permeabiHty of jellies, also increase the disinfectant activity of many antiseptics. In practice the conditions are compHcated enormously. We are no longer concerned with the distribution of the disinfectant between solvent and microorganisms but organic substances are added (sputum, albmnin, feces) so that we have the sums of unknown factors which can only occasionally be resolved. The action of a disinfectant is usually much depressed by organic matter. This is also the reason why disinfection of the organism, an internal disinfec- tion or antisepsis, has so seldom been accomplished by chemical means. There are, indeed, substances so slightly toxic, that men or 404 COLLOIDS IN BIOLOGY AND MEDICINE animals may take the dose theoretically necessary to disinfect the body, for instance, tetrabrom-o-cresol and hexabromdioxydi- phenylcarbinol which, according to H. Bechhold and Ehrlich, stop the development of diphtheria bacilli in bouillon at a dilution of 1 : 200,000. In the organism they have no effect at all, in spite of the fact that there may be introduced into the body without harm, doses which are one hundred times that necessary to inhibit the development of the bacteria in vitro or to kill them within twenty- four hours. Tetrachlor-o-diphenol behaves similarly; it inhibits development of diphtheria bacilli in dilutions of 1 : 400,000 to 1 : 640,000. Individual colonies still grew in a serum culture in the presence of the chemical at a dilution of 1 : 10,000. We might question whether the result was due to favorable vital conditions in serum removed from a living organism or to other causes. Experi- ment proved the latter view correct. By ultrafiltration the free tetrachlor-o-diphenol was separated from the fraction bound to serum colloids and it was found that 87.5 per cent of the disinfectant had been fixed by the serum colloids. The relatively simple conditions in the disinfection of skin and hands are especially instructive. The hands adsorb soHd particles from the air and particles of dirt and bacteria from dirty water (H. Bech- hold). Upon washing with soap these particles are surrounded by fatty acids or fatty acid alkali hydrolytically split off and cease to cling to the hands. A priori we might conclude that there would be a diminution in germs or disinfection associated with the cleaning of the hands; indeed, it was shown by earher investigators and recently by H. Reichenbach that soaps possess considerable germ-killing action. I was able to prove that there exists absolute parallelism between the detergent and the disinfecting action of soaps. It is impos- sible to disinfect the hands with soap in any practicable time (10 minutes), though this can be readily accomplished with alcohol and alcohohc solutions. According to H. Bechhold, the reason is that alcohol with its low dynamic surface tension readily enters the capillary interspaces of the bare hand where the bacteria are lodged, but aqueous solutions, on the contrary, enter them very slowly. This can be readily discovered by the difference in the distance they ascend in strips of filter paper. [As the result of trench warfare the study of antiseptics in the treatment of wounds has received intensive study. Antiseptic sur- gery has been revived. Carrel and Dehelly have elaborated a valuable system for the treatment of wounds by irrigation with anti- septics of the chlorin group. The whole subject of ;wound irrigation has been restudied and new antiseptics discovered. TOXICOLOGY AND PHARMACOLOGY 405 J. F. McClendon, in the Journal of Laboratory and Clinical Medi- cine, August, 1917, discusses "The Relation of Physical Chemistry to the Irrigation of Wounds." He emphasizes the importance of protecting the tissues from the effects of prolonged diffusion. The action of the antiseptics employed is oxidative. "Oxidizing sub- stances are, however, reduced by cells and an ideal local antiseptic would be one whose reduction product is indifferent. Hydrogen peroxide falls in this class but is not a powerful oxidizing agent and is decomposed by catalase so rapidly as to render a large per- centage of it ineffective. It acts as a mechanical cleanser. If infusoria are placed in a solution of H2O2, the latter penetrates their protoplasm and is decomposed on the inside with the Hberation of bubbles of oxygen which burst and destroy the cells. More useful agents are iodin and chlorin, especially the latter since HCl formed on its reduction may be neutrahzed by NaHCOs that has been added, and thus rendered indifferent. According to Dakin and his col- laborators, chlorin forms chloramines when it acts on protoplasm, and these chloramines have an antiseptic action. It is true, however, that chlorin oxidizes many organic compounds with the liberation of HCl. Chlorin gas escapes rapidly from its solution in water, but this may be retarded by the addition of a base transforming it into hypochlorite. Its oxidizing power is impaired, however, if the reaction is very alkahne, but may be restored by bubbling CO2 through the solution." McClendon emphasizes the importance of having the irrigating fluid physiologically normal. It is not enough in his opinion that the solution should contain the salts in the proper proportion but it must have the correct hydrogen ion concentration (Ph)- This may be provided by bubbhng CO2 through the fluid and measuring the p^ with indicators. The most important new antiseptics are chloramine T or sodium toluene sulphonchloramide soluble in water, and dichloramine T or toluene-p-sulphone dichloramine soluble in organic solvents, and a paraffin saturated with chlorin, called chlorocozane. See Handbook of Antiseptics (Dakin and Dunham). Tr.] The Method of Testing Disinfectants Considered in the Light of Colloid Research. For testing disinfectants bacteria are usually dried on silk threads or marbles. These are dipped in the disinfectant solution, and after the solution is removed they are placed in bouillon or fluidified agar. If the bacteria have been killed by the immersion, no germs develop. 406 COLLOIDS IN BIOLOGY AND MEDICINE From the length of time required to kill the germs and from the con- centration of the disinfectant solution we may judge the strength of the disinfectant action. From the standpoint of the colloid chemist the silk thread pro- cedure contains a serious error of method. Even at present on account of its apparent simplicity this method is frequently em- ployed. We know from practical experience that silk is a very powerful adsorbent. The investigations of W. Schellens * on the relation of silk to sublimate is of interest in this connection. He shook 1 gm. silk with 50 c.c. of 1 per cent sublimate solution and then determined how much mercury was present both in the re- maining fluid, and in the silk after it had been washed many times. He found that the silk had taken up 6.04 per cent of its weight of metallic mercury but had fixed only 1.9 per cent. We thus see that silk retains very considerable quantities of sublimate. Similar re- sults were obtained by W. Schellens for ferric chlorid, ferric acetate, several mercuric salts, lead nitrate, etc. From this we must con- clude that silk is not a suitable germ carrier for disinfection experi- ments, since as the result of adsorption (no action can be ascribed to the "fixed" mercury, etc.) it retains too much disinfectant; on this account the germ cannot escape from the disinfectant, and ac- cordingly we are only given information relative to inhibition of development and not concerning the lethal action. Paul and Krönig chose, as germ carriers, marbles because the disinfectant can barely adhere to them by adsorption. H. Bechhold and P. Ehrlich,* as well as H. Bechhold, *9 in their experiments on lethal action completely discarded germ carriers; they prepared bac- terial cultures on agar, which they covered with the disinfectant fluid. After removing the disinfectant, they washed the culture twice with physiological salt solution (which is finally made very faintly alkaline) and then transplanted the culture to a new medium (agar). On account of the thickness of the culture, very great demands are made upon the disinfectant by this method, but no germ carrier whatever is transferred to the new culture medium, and the method thus completely avoids the source of error men- tioned above. The experiments on the disinfectant action of formaldehyd gave such contradictory results, because the great adsorption of formalde- hyd by silk was ignored, as was pointed out, especially by Schum- BERG.* In order to annul the adsorptive action of germs and germ carriers in disinfection experiments, an attempt was made to render the disinfectant inactive by chemical means, as it was found impossible TOXICOLOGY AND PHARMACOLOGY 407 to accomplish this by washmg. J. Geppert * inactivated sublimate by means of the action of ammonium sulphid ; the sublimate is thus changed to the innocuous mercuric sulphid. In the case of formal- dehyd, ammonia is employed, for by means of ammonia, formalde- hyd is changed to hexamethylentetramin. There is no chemical agent destructive for phenol and phenol-like compounds to which objections cannot be raised. From the colloid-chemical standpoint, I regard the principle of chemically removing the disinfectant as erroneous in many cases. The idea which guided Geppert and his successors was evidently that if a germ which had been immersed in a disinfectant is placed on a suitable culture medium, the medium abstracts the last traces of the adherent disinfectant; it is thus washed just as a chemist washes a crystalline precipitate on a filter. In this way we consider the effect only for the time during which the germ remained in the disin- fectant, and J. Geppert and his followers seek to imitate this limited time by chemical destruction of the disinfectant when the germ is removed. As a matter of fact the process proceeds differently: when the germ is removed from the disinfectant and is placed on a fresh culture medium, it releases the disinfectant only slowly and incompletely in accordance with the laws of adsorption. We may compare the process to the ''bleeding" of dyed fabric; especially the bleeding of cotton which has been dyed with a dye that is chemi- cally insufl&ciently fixed by the fiber and which for days gives up color when washed with water; the dyer says it '' bleeds." Thus for a long time by a pure adsorptive action the germ retains the disinfectant and is injured by it. That this assumption is correct is shown by some experimental results taken from the literature, throughout which the expression is employed that the germs are "weakened." This expression appears to me to be the transfer to organisms to which it no longer applies of a conception applicable to men and higher animals. According to J. Geppert, anthrax spores are weakened but not killed by the action of 0.1 per cent sublimate solution for 15 min- utes. They are unable to develop even in a culture medium which contains as little as 1 : 2,000,000 sublimate, whereas normal an- thrax bacilli thrive quite well in this medium. Our interpretation of this is that anthrax spores previously treated with 1 : 1000 sublimate adsorbs so much sublimate that they are in adsorption equilibrium with a nutrient medium that contains 1:2,000,000 sublimate. Heinz says, "Sublimate acts in animal infections just the same as when transplanted upon artificial media and the minutest traces 408 COLLOIDS IN BIOLOGY AND MEDICINE suffice to prevent multiplication or the infection of animals on the part of the germs weakened by the antiseptic. " ''Anthrax bacilli (Heinz) hke anthrax spores prior to the lethal action show a stage of weakness in which the bacilli are unable to grow in a nutritive medium containing a minimal amount of disin- fectant. Thus anthrax bacilli which had been immersed in 1 per cent carbolic acid (and had not been killed) did not grow in a culture medium which contained a small amount of carbolic acid, " whereas fresh anthrax bacilli grew luxuriantly. I find a very instructive example in Ottolenghi's * paper. He says, ''The fact is very interesting, that occasionally certain paper strips (he soaked blotting paper strips with an emulsion of anthrax spores, dried them and then placed them in sublimate solution) after they have been subjected for 24 hours to a sublimate solution (up to 2.712 per cent) and were inoculated into guinea pigs, may yield a luxuriant development of anthrax bacilli if they are removed from the thoroughly healthy animal after one week and are placed on media after a thorough treatment with H2S. " The results of H. Reichen- bach ^ are to be judged from the same standpoint. After treating anthrax spores with sublimate, they first lost their activity in the bodies of animals, then their ability to grow in bouillon (without ammonium sulphid treatment) and only after a much longer time did they cease to grow, even after treatment with ammonium sulphid. Unquestionably numerous analogous examples would be found were the literature carefully studied. It may be seen from this that in disinfection experiments, the chemical removal of the disinfectant may lead to false results, that it may simulate a weaker action of the disinfectant than it actually has, i.e., a weaker action than it possesses in practice under natural conditions. On this account I regard repeated washing of the germs with indifferent solvents (water or physiological salt solution which is finally made faintly alkaHne with soda) as the proper method for the removal of the disinfectant. Whatever is retained by the germ after such a washing would also be retained under natural conditions. The Bechhold-Ehrlich method of killing germs (see p. 406) meets all these conditions correctly. This criticism relates to the testing of disinfectants against germs which can directly enter the organism (disinfection of the hands, antiseptics, etc.). It is otherwise with substances which serve for the disinfection of stools, sputum, etc. Under these circumstances we must consider that the disinfectants penetrate an environ- 1 According to personal letter. See also H. Reichenbach, Zeitschr. f. Hy- giene und. Infectionskrankh., 50, 455, u. 460-462 (1905). TOXICOLOGY AND PHARMACOLOGY 409 ment which contains hydrogen sulphid, ammonia, etc. The testing of a disinfectant must always take its use into consideration and be accordingly varied in different cases. [The criterion of Carrel and Dehelly is the bacterial count per field in smears taken from the wound. Tr.] Diuretics and Purgatives. Diuresis and defecation may be influenced in the most varied ways, for instance by increased blood pressure or by increased peristalsis — in brief, by such factors as chiefly exert a more or less specific nervous action; similar effects may be obtained by a purely mechanical facilitation of secretion or by hindrance of absorption. We have repeatedly referred to the lyotropic series of the alkaline salts (see pp. 80 and 296) and have shown among other things, that there exists a remarkable parallelism between the swelling of gelatin and fibrin, the precipitation of albumin and lecithin and the irrita- bility of frog's muscle and ciliated epithefium. Also for diuresis and defecation there exist such evident relationships which we shall here elucidate. We give the classification of F. Hofmeister. The figures above the columns I, II, etc., indicate the concentration of the salt solutions which are necessary to salt out globulin. I. II. III. IV. V. 1.51-1.66 2-2.03 2.51-2.72 3.53-3.63 5.42-5.52 Li sulphate Na sulphate Na phosphate K phosphate K acetate NH4 sulphate Mg sulphate NH4 phosphate NH4 citrate NH4 tartrate Na carbonate NaCl KCl Na nitrate Na chlorate Na acetate K citrate Na citrate K tartrate Na tartrate The various members of Group I are purgatives; those of IV and V are diuretics, while the action of those in II and III with the exception of magnesium sulphate are not sufficiently definite to be of any service. Obviously, the anion is of the greatest importance for the action of the above salts: we observe that CI and NO3 have the highest rate of diffusion and are most rapidly absorbed. NaCl, KCl and NaNOs aid sweHing so that a gel swells more rapidly in such a salt solution than in pure water. From this it follows that the in- 410 COLLOIDS IN BIOLOGY AND MEDICINE testine will take up such solutions more rapidly than pure water. Accordingly, all the conditions necessary to give the body a large quantity of dilute salt solution are fulfilled. We know from Chap- ter XIV that there is a strong effort on the part of the manmaalian organism to keep constant the swollen condition of the blood and tissues, as well as the osmotic pressure. For this purpose, the kidney is most important, since it is able to remove excess of water and salts. We may even at present recognize thus the quahtative relationship between physical properties and the diuretic action of Groups IV and V. Unfortunately, we are not in a position to pursue the process quantitatively, but we may assume that there would not be a simple relationship. The above-mentioned physical properties of Groups IV and V are to be classified not only in reference to the intestinal membranes and the kidney function, but they also pay a role in the irritation of nerves and the contraction of muscle (see p. 289 et seq. and p. 354). According to Wo. Pauli *^ the majority of cations raise the blood pressure, whereas Br depresses it. This explains why bromids are of no use as diuretics in spite of the fact that they might be classified as such from their behavior with colloids; the depression of blood pressure they cause opposes their diuretic action. Hypotonic common salt solution and potassium nitrate solutions remain therefore the chief diuretics among the alkali salts. In fact, it is the solution of common salt which plays, in the Spa "mineral water cures," the chief part in increasing the urinary excretion. The result is quite different when solutions are introduced directly into the blood stream. A physiological salt solution is excreted practically quantitatively. If we inject a hypertonic salt solution, then more water will be excreted than was introduced, and (within certain limits) proportionately more will be excreted the greater the concentration. This is not surprising, because the salt withdraws the water of swelling, especially from the blood corpuscles and the muscles. The water thus set "free" is then filtered away by the kidneys. Sulphates, phosphates, tartrates and citrates, etc., of sodium impede diuresis when taken by mouth; however, when di- rectly injected into the blood stream, they are even more strongly diuretic than common salt. This depends on their strong dehydrat- ing action and their low diffusibihty. Martin H. Fischer by in- troducing such salts was able to make a kidney, which had been edematous by ligating the renal artery, function again. On in- jecting an appropriate salt into the renal artery or even into the kidney itself the swelling subsided and the anuria ceased. According to E. Frey,* if we inject the salts mentioned along TOXICOLOGY AND PHARMACOLOGY 411 with narcotics (morphine, chloral, ether, urethan), no diuresis de- velops, and on the other hand the absorption of water from the in- testines is unimpaired. This is explained by the fact mentioned (p. 338), that such narcotics inhibit the oxidizing processes in the organism, which results in a greater fixation of water ("acid swell- ing"). What holds for electrolytes is also true for nonelectrolytes. We have recognized in urea a substance which greatly aids diffusion through jellies (see p. 55) and which opens through the hydrogel paths for itself and other substances; as a matter of fact it acts as a diuretic. I wish to mention some additional facts concerning ammonium salts and the cleavage products of protein. All the evi- dence (see pp. 80 to 82) is in favor of the view that the action of the cations and of the anions of an electrolyte is antagonistic and that they mutually counteract a portion of their own activity. Thus NH4 seems to oppose the precipitating and dehydrating action of SO4, citrate, and tartrate anions to a greater extent than K and Na (see the Series III of our group). If we bring this into relation with /NH2 analogous action of urea CO m > ^^ o x) OJ O fl d o fl > G, ^ > ^ -►^-^-'"S c3 ^ > ■S'3.2 CO a 03 Sh PhP>h> PhI-:! P fepq - QfQ p PrW rt rt > ^^ a^ p o rt o bC^ a> ra _, P^ Q p^ (^ (^ .1-1 n3 d a & <0 QJ PQ p fiHpQ fcj Q Ü H cd . — 1 'O o m tH .. > > at. bU d at he blue ark aint aint at he O rt P finf^P^ TS d £3 d-ö ^ d -a ^ d t-i aa >>Ü ^P= d a» a; Sh t^ tH O T3 4) OJ a; 'Z! >PQ rt pR d o d o d o rd bD d o _d 3 OJ ^ ^ ^ MOM .2-^ k^ o oä-d |J3 2.S w .2 ^. oS 'S ^ a' o a; bC ;3 d «2 o "^ d -e o dT3 cC -d a o o _^ — - CD % «3 a:'-:3 tu "2 -.5 5 t< fe. fe o d o) (u ffS^ °<^ a d t3 oo d^ ■^t CO . — I Ö «CT QQ >^ CO d . <1M .S -►^■':5 •" CO TO d «3 dcß O t^ +*. 5-; ■^-' q; Ol tu >>2 & T3 c3 d d -t^ r-« r3 rd o3 .S o d a> 03 CO^ ■ — 3 03 d ^"ö k-5 *-' (B d c3 rd'M 2 OM<1 -^ d d . ^ OJ o :2^ CD « S 03 ÜPQ d d Co 1-5 pa 430 COLLOIDS IN BIOLOGY AND MEDICINE dyed directly only by certain groups of dyes. On this account he is able to obtain almost uniform dyeing. The biologist, on the con- trary, has to stain much more numerous varieties of tissue and it is quite wonderful what different shadings and even what different colors the individual tissue elements of a tissue often show after treatment with a single stain. If we consider, moreover, that each individual tissue element frequently selects its own particular constituent from a mixture of stains, thus forming the brightest pictures, we shall understand why the chemical theory of dyeing is the most convinc- ing to biologists. The following is a resume of the results to date of the investiga- tions upon dyeing with basic and with acid dyes. By adsorption the dye is concentrated upon the tissues, with which a fixation may occur as the result of chemical processes. Hitherto we have only considered the so-called substantive dyeing, by which the fabric stains directly in the dye solution without any previous preparation (wool, silk). Vegetable fibers, such as cotton, flax, paper, etc., take up very little color from most dye solutions and do not hold it very firmly; they require a mediator to chain the color to them, namely a mordant. This type of dyeing is known as adjective dyeing, a term introduced into industrial dyeing tech- nology by J. BANCROF.T. In biological staining the chief mordants are alum and ferric oxid salts. The combination of mordants with dyes (hemotoxylin, hematin and alizarin colors) are called lakes. W. BiLTZ has definitely proved that in addition to physical adsorp- tion, chemical combinations occur in adjective dyeing. Histologically, staining with dye mixtures is much employed. P. Ehrlich found that if aqueous solutions of an acid dye, e.g., acid fuchsin or orange G, were mixed with a basic one, e.g., methylene blue or methylene green, so that one remained in excess, then no precipitate was formed. From colloidal solutions of dye mixtures certain tissue elements remove the basic and others the acid dye. It is thus possible to obtain with one solution double stains, or even triple stains (triacid). (See above.) According to the investigations of 0. Teague and B. H. Buxton, *2 acid and basic dyes precipitate most completely if they are mixed in equimolecular proportions. An excess of one dye interferes with precipitation, i.e., it acts as a protective colloid, and, in fact, the in- terference zones are wider, the more colloidal the dye. Especially important for the histologist is the fact that highly colloidal dye mixtures are bound more firmly together than those that are slightly colloidal. We must consider very critically, microchemical reactions and MICROSCOPICAL TECHNIC 431 stains which arise from the interaction of two chemical substances with the formation of an insoluble precipitate. R. Liesegang*^ has called attention to the phenomena involved in his investigations of Golgi's stain. If a piece of brain is placed in potassium bichro- mate, and after it is completely soaked through, it is then immersed in silver nitrate, some of the ganglion cells in which, silver Chromate has been precipitated are stained reddish brown. The interior of the brain substance is never thoroughly stained, notwithstanding the fact that potassium bichromate is present after the first process and silver nitrate after the silver bath. The reason is as follows: when the chromatized portion of brain is placed in the silver nitrate solution, silver Chromate forms in the outer layers; the potassium bichromate present in the interior diffuses outward where it is arrested by the silver so that the interior is more and more depleted of Chromate. The irregularities in the Golgi staining de- pend upon similar interferences with diffusion and nucleus actions of silver Chromate, on account of which only a portion of the ganglion cells are stained. After staining peripheral nerves with Golgi's stain we obtain stratifications in the axis cylinders (Fromann's lines). These have been shown to be artifacts. The Technic of Staining. We distinguish staining en masse, section staining and vital staining. In staining en masse the entire object is immersed in the stain solution subsequent to hardening. If this is soluble in alcohol, it requires no special precautions; it is otherwise with solutions con- taining alum, in which case alcohol in the object must first be re- placed by water. After staining, the dye is washed away with water or alcohol until the fluid remains colorless. After staining in aqueous solution the piece must be rehardened in alcohol. The subsequent treatment is then the same as in unstained pieces. Section staining is much more frequently employed, since not only details are brought out better, but the staining can be watched more closely and later counter-stains may be added intermittently. Ac- cording to the dilution of the stain solution and the length of time the section is stained, we may obtain on the one hand contrasting, or on the other finely shaded pictures with much more detail. Vital staining, the staining of living tissues, was introduced by P. Ehrlich and was apphed by this investigator in his classical work on "The Oxygen Requirements of the Organism" to the processes of living cells. At present it has the center of interest, and from it we 432 COLLOIDS IN BIOLOGY AND MEDICINE may expect most valuable discoveries on the physiology and pathology of living tissue as well as concerning the mechanism of the action of drugs. E. GoLDMANN, R, HÖBER and W. Schuleman have in recent years contributed much concerning the utiHzation and theory of vital staining. They studied healthy and sick animals, whereas Küster and RuHLAND apphed vital staining to plants. As yet vital stain- ing of bacteria and other microorganisms has not been definitely attained (Eisenberg). The stain must not be poisonous or the cell will die before it has the desired color.^ We have numerous dyes at present which fulfill this condition. A few of the most useful are mentioned, methylene blue, neutral red, toluidin blue, trypan blue, trypan red and isamin blue. The studies of Ruhland on plants, as well as those of Evans, Schuleman and Wilborn on animals, indicate strongly that the extent of dispersion of the dye chiefly determines its suitability for vital staining, so that the cell behaves like an utra- filter (see p. 428). A dye that is too diffusible distributes itself too readily in all the organs and is accordingly quickly excreted by them; one that is highly colloidal remains at the site of injection. The studies of Ruhland include both basic and acid dyes while the experiments of Evans, Shulemann and Wilborn were only with acid dyes. It was formerly believed that only lipoid soluble dyes penetrated living tissues, but this view has not been sustained (see also Garmus) . Many vital stains are known which are insoluble in fats. The col- loidal metals are included among these; they have proven useful agents in studying "distribution" in J. Voigt's method of investiga- tion. This does not by any means imply that lipoid insoluble vital stains may not be especially suitable for some of the organs which are rich in hpoids. Thus, for instance, axis cylinder and ganghon cells of the nerve substance are most intensely stained by methylene blue. It is remarkable that the cell nucleus which stains most intensely with basic dyes when the object is dead, with vital staining is con- stantly colorless; nuclear staining occurs only when the cell dies. If vital stains are to be fixed, i.e., made insoluble, ammonium molybdate, sublimate, picric acid, etc., are employed. If this fixation is omitted, the dye diffuses away after death, i.e., according to the changed condition of the tissue, physical and chemical, and a different distribution results* 1 The lack of toxicity of vital stains is only relative; in concentrated solution they are all poisons and may be used only in extreme dilutions. Safranin and methyl violet, especially, are quite poisonous, and on this account they cannot be employed for injections into the higher animals. MICROSCOPICAL TECHNIC 433 THE TISSUE ELEMENTS IN THEIR RELATION TO FIXATIVES AND DYES. , With iodin-potassium iodicl solution, starch grains give a blue adsorption compound (see p. 135). Glycogen forms with it a red adsorption compound. The stain with strongly alkaline potassium carmine recently recommended by Best is so complicated that it cannot yet be interpreted. The Lipoids. The fixation and staining of lipoids can hardly be regarded as other than a colloid-chemical question. Fixation is generally ac- complished with osmic acid by means of which the fat is simultane- ously blackened and the acid is reduced to colloidal metallic osmium; similarly, gold, silver and palladium salts are reduced to the colloidal metal. Of the true dyes, we must especially consider those which are very soluble in fat, though quite indifferent chemically, and which are very slightly adsorbed by the other constituents of the cell. Among these are Scarlet R (fettponceau) and Sudan III. Both are amphoteric dyes in which the basic as well as the acid character is so indefinite that they seem quite indifferent and do not form salts with aqueous caustic soda. Employed in alcoholic solution, staining results. Protoplasm. We may attribute to protoplasm chemical properties similar to those of the albumins. Protoplasm may be amphoteric, on which account neither acid nor basic properties become more prominent. Consequently, protoplasm stains only faintly with either basic or acid dyes, even though its water content is relatively high. Nucleus. The chief constituents of the cell nucleus are the nucleoprofeins. These are strongly acid in character; to them may be attributed the intense staining of the nucleus with basic dyes, and to them the in- tensely staining constituent of the nucleus is indebted for the name chromatin or chromatic substance among histologists. The union Tvdth the color base becomes firmer with the lapse of time, since in the beginning it is possible to effect almost complete decolorization A^ith alcohol, whereas when the dye acts for a longer period the nuclei retain their intense staining and only clouds of color leave. For nuclear staining any basic dye may be employed; safranin, fuchsin, methyl violet, methyl green and bismarck brown are recommended most highly. 434 COLLOIDS IN BIOLOGY AND MEDICINE Another favorite nuclear staining method is with the mordant dyes, e.g., hematoxyhn or carmine. In this instance, also, the acid char- acter of the nuclear proteins explains the action of the dyes. The nuclear proteins adsorb the mordants, usually colloidal aluminium hydroxid (from alum), and these form an insoluble compound with the acid hematoxylin or one of its oxidation compounds, or with acid carmine. Finally, we may mention the double staining of Romanowsky, which has been modified by G. Giemsa. Its underlying principle is that a basic blue dye (methylene azur or methylene blue) is mixed with an acid dye eosin (see p. 426). At first the preparation stains blue in the mixture; gradually there occurs a differentiation into blue and red elements or combination violet shades whereby the nuclei become red. For the present, all interpretations of this phe- nomenon are quite hypothetical ; it presents a very interesting colloid- chemical problem. If methylene azur and eosin are mixed, a colloidal solution of eosin-acid-methylene azur forms, provided that one of the two dyes is present in excess. Nuclear staining may occur in such a way that the basic methylene azur serves as mordant for the eosin; it is also possible that the nuclei stain better with colloidal eosin- acid-methylene azur than with crystalloidal methylene azur, and that in a reaction which requires time (possibly hydrolytic cleavage) the red color base of methylene azur becomes free. In this double staining there enter as factors phenomena involving the colloidal condition of both dye and specimen with respect to the diffusibility of the dye and perhaps also other circumstances which have not been consid- ered here. This may be assumed both from the accurate directions which are given for the preparation and age of the solution, the thickness of the preparation, the duration of staining, etc., and from the fact that every departure from the directions gives a dif- ferent result. Connective Tissue, Capillary Walls, Membranes, Etc. From the numerous reports I gather that only easily diffusible stains, especially the sulphoacids (acid fuchsin, soluble blue com- bined with picric acid), are suitable for this purpose. This probably depends upon the fact that connective tissue, etc., are among the tissues poorest in water and least swollen, so that dyes of more col- loidal character are unable to penetrate them. For the staining of elastic fibers, which is best performed by the orcein method of P. G. Unna and Taenzer or by Weigert's method, we have no explanation whatever. The recent investigations of the keratins by L. Golodetz and P. G. Unna show that we are dealing MICROSCOPICAL TECHNIC 435 with a number of chemically very different substances (ovokeratin, neurokeratin, elastin). [Van Gibson's stain contains picric acid and stains elastin specifically. Tr.] The Staining of Bacteria. Most cocci and bacteria have a definite acid character evidenced by the fact that they migrate to the anode in the electrical current (see p. 205). Though they usually stain intensely with basic dyes (f uchsin, meth- ylene blue, thionin, etc.), nevertheless bacteria exhibit considerable differences in staining capacity. Though all cocci with which I am acquainted stain very intensely, some bacteria, e.g., paratyphoid and bacilli of hog erysipelas, are stained more faintly. Spores stain with especial difficulty, the more poorly the older they are; it is obvious that the solid capsule offers great resistance to the penetration of the stain. The tubercle bacillus is most difficult to stain, which may be attributed chiefly to its high keratin content, inasmuch as other keratin-containing substances (bristles, hair, epidermis, etc.) stain just" as poorly. The difficulty in staining the tubercle bacillus was formerly attributed to the wax contained. Helbig, however, showed that complete removal of the wax did not increase the stain- ing capacity. Gram's stain is quite unique; it is extensively employed for the classification of bacteria (we distinguish Gram-positive and Gram- negative). It is performed as follows: we first stain with methyl violet or some related basic dye and then subject the specimen to the action of iodin (dissolved in KI). After this treatment, some bac- teria readily give up the dye to alcohol and are decolorized, whereas others firmly retain it. In the latter case, a firm combination has been formed. A thorough study from modern points of view would be of great value, since it would explain the difference in the nature of the two groups of bacteria. It is important to mention that, by Gram's method, a differentiation of the structm-e of individual bac- teria may be revealed. The so-called Babe's corpuscles are not de- colorized by a brief action of alcohol. Upon this fact depends M. Neisser's method for identifying diphtheria bacilli. We have as yet little insight into the actual basis of differentiation by Gram's stain. It has actually only been established that Gram positive bacteria show a greater permeability for dies, stain more quickly and intensely and retain the dye more strongly upon de- colorization with alcohol. Probably the onlj^ purpose of the treat- ment mth iodin is to increase the size of the dye molecule or increase its fixation by the bacillus (Eisenberg) . > AUTHORS' INDEX ^^ refers to footnote in text * refers to reference given Abbe, 124, 127 Abderhalden, E., 32, 34, 72*^, 92, 187, 210, 210^», 220*1, 321 1) Lehrbuch d. physiol. Chemie (Ber- hn, 1906) 2) Zeitschr. f . physiol. Chemie 37, 484 (1903) and Guggenheim, M., 34*, 189* Zeitschr. f. physiol. Chemie 54 (1908) and Pettibone, 185* and Strauch, F. W., 187 Zeitschr. f. physiol. Chemie 71, 315- 318 (1911) Abramow, S., 206 AcHARD and Weill, E., 371 Adam, P., 311 Adie, 57* Journ. Chem. Soc. 344 (1891) Adler, H. M., 363 Journ. of Am. Assoc. 2, 752 (1908) and Herzog, B., 27* and Ringer, 379 Aggazzotti, a. and Foa, C, 374*, 384* Albarran, 340 Albrecht, E., 305* Verh. d. D. pathol. Ges. 5, 2 (1903) Albü, a. and Neuberg, C, 218*, 219*, 233* Physiol, u. Pathol, d. Mineralstoff- wechsels (BerHn, 1906) Alexander, Jerome, 42-''", 71, 174, 188, 195^" and Bullowa, J. G. M., 174, 349* Archives of Pediatrics (N. Y., Jan., 1910) (and Zsigmondy), 247 Altmann and Sachs, H., 206, 208 Amann, J., 343* BuU. Soc. Vaudoise, 38, 131 Ambard, 336 Amberger, C, 11, 366 Ambronn, H., 259* KolL Zeitschr. 6, 222-225 (1910) Ames and Bauer, J., 231, 252 Anderson, 10 Ariens and Kappers, C. U., 261* Folia Neuro-Biologica 1, No. 244 (1908) Araki, F. and Zillessen, H., 226 Arrhenius, Sv., 26*, 53, 56*, 105*, 196, 197, 200*, 213, 386 Immunochemie (Leipzig, 1907) and Herzog, R. O., 47 Arric, Le Fevre de, 371, 372 AscHERSON, 35*, 347, 347* Arch. f. Anatom, u. Physiol. 53 (1840) AscoLi, M., 109, 211, 371*1, 372* Koll. Zeitschr. 5, 186 (1909); 6, 293- 298 (1910) and Izar, G., 211, 365*, 366, 371, 373*, 377*, 383^" 1) Berliner Klin. Wochenschr. 4 and 21 (1907) 2) Biochem. Zeitschr. 5, 394; 6, 192; 7, 143; 10, 356; 14, 491; 17, 361 (1907-1909) 3) BoU. della Soc. Med. Chir. di Pavia 35 (1908) *) Comptes rend, de la Soc. de Biologie 65, 59 and 426 AsHER, L., 324* Biochem. Zeitschr. 14 (1908) AuER, 413 Amer. Journ. of Physiol. 17 (1906) and Journ. of Biol. Chem. 4 (1908) Auerbach and Pick, 330 Babcock, S. M. and Russel, H. L., 175* Ber. Landw. Ver. Stat. d. Univ. Wisconsin (Ver. St. v. N. A.) 20 437 438 AUTHORS' INDEX Babe, 435 Bachman, 9*, 10 Bachmann, L. and Rünnstrom, J., 265, 265* Biochem. Zeitschr. 22, 290 (1909) Balyer, a. von, 27 Baisch and Landwehr, 343 Bancroft, J., 413*, 430 W. D., 37, 40, 46 W. D. and Clowes, G. H. A., 36-^" Bang, J., 139, 244* Biochem. Zeitschr. 16, 255 (1909) Barbieri and Carbone, 352 Barcroft, J. 310 1) Journ. of biol. Chemistry 3, 191 (1907); Pflüger's Arch. 122, 616 (1908) and Brodie, 334 Journ. of Physiol. 32, 18, 33, 52 Bary, de and Stahl, E., 286 Battelli and Stern, 386 Bauer, J., 231, 252 and Ames, 231, 252 Bayliss, W. M., 29, 137, 166, 182, 183, 365 and Fischer, M. H., 220^" Bechhold, H., 6, 10, 11, 16, 17, 26, 29 35, 36*1, 42^ 55*2^ 55, 58*^, 80, 83 84*1, 86, 95, 95*S 95^", 96^ "-*S 96^"- 97, 97^"-, 98, 99, 99-^"-*«, 100-^"-*« 102, 102*S 102*^ 104*2, 106, 108 115, 119, 119*"-, 125, 135*S 138 144, 146, 147/", 148, 157*, 158** 164, 165, 166, 180*8, 196, 197*^ 198*4, 199*4^ 201*4, 203, 203*i 205*1, 205*1°, 241, 260, 262, 262*2 268*, 283*, 288*, 327, 332*1^, 332*" 333, 347*1, 348*4, 352/», 371 382*1, 394*^ 395-''", 396*^, 401*» 402*3, 404 1) Zeitschr. f. physikal. Chemie 48 385-423 (1904) 2) Zeitschr. f. physikal. Chemie 52 185-199 (1905) 3) Wienerklin. Wochenschr. 1905, No. 25 4) Zeitschr. f. physikal. Chemie 60, 257-318 (1907) ^) Münchner med. Wochenschr. 1908, No. 34 Bechhold, H., ^) Zeitschr. f. physikal Chemie 64, 328-342 (1908) ^) Zeitschr. f. physiol. Chemie 52, 177-180 (1907) 8) Koll. Zeitschr. 2, Nos. 1 and 2 (1907) ^) Zeitschr. f. Hygiene u. Infektion- skrankh. 64, 113-142 (1909) 1") Münchner med. Wochenschr. 1907, No. 39 11) Koll. Zeitschr. 5, 22-25 (1909) 12) van Bemmelen-Festschrift (1910) and Ehrlich, P., 394*, 396, 404, 408 Zeitschr. f. physiol. Chemie 47, 173- 199 (1906) and Ziegler, 10*, 35, 54*^, 55, 55*2, 57*1, 73^ 106, 138*2, 147 /«^ 143*3^ 162*2, 238*1, 243, 244, 260*i, 269*% 319, 320*2, 327, 329, 343, 378*% 380, 403*2 1) Ann. d. Phys. (4) 20, 900-918 (1906) 2) Zeitschr. f. physikal. Chemie 56, 105-121 (1906) 3) Biochem. Zeitschr. 20, 189-214 (1909) 4) Berliner klin. Wochenscher. (1910) Beck and Hirch, 113 Beckmann, E., 115 Behring, E. von, 144 Beitr. z. exper. Therapie 10 (1905) Beijerinck, M. W., 76 Bemmelen, J. M. VAN, 9, 26, 67, 152-''", 217, 329, 337 Bence, J. and Koränyi, A, von, 315 Benedicenti and Revello-Alves, 157 Benedict, H., 312* Pflüger's Arch. 115, 106 (1906) Beniasch and Michaelis, 204 Bentner, G., 230 R. and Loeb, J., 240, 295, 295^" Berczeller, 143* and Czäki, 29* Berg, W. and Fischer, A., 418 Berghaus, W., 359 Berman, 17 Bernoulli, E., 381 Bernstein, 296 E. P. and Simons, Irving E., 211^'' Berthold, G., 283* Studien über Protoplasmamechanik (Leipzig, 1886) AUTHORS' INDEX 439 Bertholet, M. and Jungfleisch, 20 Bertrand, G., 191 Best, 433 Betegh, von, 102 Bezold, a. von, 265 BiERRY and Henri, V., 191* Comptes rend, de la Soc. de Biologie 60, 479 (1906) Billiter, J., 86* Zeitschr. f. physikal. Chemie 51, 142 (1905) BiLTz, W., 26*1, 28, 31*S 47, 86, 106, 107, 110, 125, 135, 135*, 145*^ 196, 197, 200*3, 384*2, 430 2) Ber. d. deutch chem. Ges. 37, 3138 (1904) 3) Zeitschr. f. Elektrochemie, 1904, No. 51 ') Biochem. Zeitschr. 23 (1909) and Freundlich, H., 110 and Gatin-Gruszewska. L., 73 Pflüger's Arch. 105, 115 (1904) Much, H. and Siebert, C, 196, 198*, 199*, 204* in E. von Behring's Beitr. z. exper. Therapie 10 and Steiner, H. 28* Koll Zeitschr. 7, 113 (1910) and Vezesack, A. von 46, 73*, 106*, 106 '"•, 237* 290* Zeitschr. f. physikal. Chemie 66, 68, 357-382 (1909); 73, 481-512 Bischoff, E., 215^", 218*, 219* Zeitschr. f. ration. Medizin 20, 75-118 (1863). Blasel and Matuta, J., 156''^'' Blom, 184* Bloor, 120 Blunschly, H., 310 BoBERTAG, O. and Feist Fischer, H. W., 67* BODENSTEIN, M., 187 and Dietz, 187 BöHi, 55* Bohr, Chr., 309 BOKORNY, 383 BoNDi, S. and Neumann, A., 247, 247*, 248-'"", 373 Wiener klin. Wochenschr. 20 (1910) BORÜET, J., 207 • Ann. de I'Inst. Pasteur, 1899, 225; 1900, 257; 1903, 161. and Gengou, 207 and Massard, 2SG BoRKowsKi and Dunin, 205* BoROWiKOW, 267, 267*, 278, 279 BosANYi and Mansfeld, 386 Bosworth, 323 BoTTAZzi, F. and D'Errxco, G., 136* Pflüger's Arch. 115, 359 (1906) P. 164, 165, 165*1, 228, 294 Rend. d. R. Acad. d. Line. 17, ser. 5" (1908); 2, sem. fasc. 9, 10; 18, ser. 5" (1909); 2, sem. fasc. 8, 9, 10; 19, ser. 5" (1910); 2, sem. fasc. 4, 210 Ph. and Onorato, 340* Arch, di Fisiol. 1, 3 (1904) BoTTERi, A. and Landsteiner, K., 199* BOURGUIGNON, 373* Compt. rend. Soc. Biol. 64 M 22, 1090 BoviE, 144 Boyle, 51 and Gay-Lussac, 51 Bredig, G., 9, 83, 99, 154*, 184, 187, 366, 371, 373, 374 1) Zeitschr. f. physikal. Chemie 51 (1898) 2) Zeitschr. f. Elektrochemie 6. 33 (1899). and Fajans, 188* Zeitschr. f. physikal. Chemie 73, 25 (1910) and Fiske, 188 and Svedberg, Th., 4 Brieger, E. and Fischer, H. W., 309 Brodie and Barcroft, J., 334 Brown, Robert, 49, 49-'^" Brownian-Zsigmondy Movement, 53, 83 Bruns, J. and Spiro, K., 401*, 403* Bruyn, L. de, 41-'^" bubanovic, 386 Büchner, 98 G. and Klatte, 184* Bugarszky, J., Roth and Steyrer, 336 and Taugl, K., 336 St. and Liebermann, L., 152*, 153* Arch, f . d. ges. Physiol. 72, 51 (1898) 440 AUTHORS' INDEX BuGLiA, G., 296 BuLLowA, J. G. M., 169^", 174, 175, 234 See also Translator and Alexander, J., 174, 349* BuNSEN, R., 363, 384 BuRiAN, R., 59*1, 102*1, 102*2, 333 1) Arch, di Fisiol. 7 (1909) (Fano- Festsch.) 2) Pflliger's Arch. d. Physiol. 136, 741-760 BuRRi, 126 BuRRiDGE, W., 234, 292, 298, 298^, 298^, 325, 354, 361 Burton, R., 84 BÜTSCHLI, O., 10, 67 Buxton, B. H., 84*^ and Rahe, A. H., 203* Journ. of med. Research 20, No. 2 (1909) Shaffer, P. and Teague, O., 203* Zeitschr. f. physikal. Chemie 57, 47-89 (1907) and Teague, O., 84*^, 430*^ Calcar, Van and Bruyn, L. de, 41/n Camerer, Jr., 265 Carbone and Barbieri, 352 Carrel, Alexis and Burrows, 246* 1) Journ. of Exper. Med. (1910-1911) 2) Berliner Idin. Wochenschr. (1911, No. 30) and Dehelly, 404, 409 Cavadias, J., 375 Cervello and Le Monaco, 362 Chabonier, 336 Chalupecky, 144 Chamberland, 187, 190 Charrin, Henri, V. and Monnier- VlNNARD, 374* Chassin, S. and Höner, R., 427* Chiari, R., 68, 161 R. and Januschke, 222 Chick, H. and Martin, C. J., 143*, 146* Journ. of Physiol. 40, 404 (1910) Chirie and Monnier-Vinnard, 374* Comptes rend, de la Soc. de Biologie 61, 673 (1906) Clausius, 50 Clowes, G. F. L., 109, 240, 241, 242, 378 G. H. A., 12, 36^", 37^", 38, 40 and Bancroft, W. D., 36-''" and Fischer, M. H., 12, 175 Coehn, a., 78* Ann. d. Physik 64, 217 (1898); 30, 777 (1909) CoNHEiM, Julius, 223 Otto, 320 CoRi, J., 417 CORNALBA, G., 346 Rev. zen. d. Lait 7, 33, 56 (1908) Cramer, 345 Crede, 365, 366 1) Apotheker-Zeitung 11, 165 2) Kongress d. D. Ges. f. Chem. Ther., Oct. ") Berliner klin. Wochenschr. No. 37 (1901) 4) Arch. f. klin. Chirur. No. 69 (1903) ^) Zeitschr. f. arztl. Fortbildung No. 20 (1904) CusHNEY, A. R., 336 and Wallace, J. B., 319* CzAKi and Berczeller, 29* Czapek, F., 240, 248 1) Ber. d. D. Botan. Ges. 28, 159-169 (1910) and Traube, J., 240 386 Dabrowski, 72, 103, 105^"-, 146* Dakin, 405 and Dunham, 405 Dam, H. van, 164* Chemisch Weekblad 7, 1013-1019 Davenport, 264* Boston Soc. Nat. Hist. 28, 73-84 (1877). Davidson and Michaelis, 202 Davis, J., 31 Dean, 207 Dekhuyser, C. and Stoeltzner, W., 420 Delaunay, 365 Demoor, J., 335 and Philippson, 298* Bull, de I'acad. de med. de Belg., 655 (1908-1909) Denham, W. S., 187* Zeitschr. f. physikal. Chemie 72, 641-694 (1910) AUTHORS' INDEX 441 Denys, G. and Leclef, 288 Determann, H. a., 113, 114, 114^"-. 310 310*, 311 Medizin. Klinik No. 27 (1910) Determeyer and Wagner, 343* Biochem. Zeitschrift, 7, 388 (1908) Devaux, 34 DiETZ, 187 Dioscorides, 364 Dender, 309 Donnan and Harris, 46 F. G., 59, 59*, 61, 62 and Donnan, W. D., 343*, 344* S., 47 W. D. and Donnan, F. G., 343*, 344* Brit. Med. Journ., Dec. 23 (1905) Dreyer, G. and Hausen, 144 and Sholto, J., 28* Proc. of Roy. Soc. 82B, 168, 185 (1910) Sholto, J. and Douglas, C., 200 DucLAUX, J., 91, 95, 102 and Malfitano, G., 92, 102 Düngern, E. von, 202* Zentral, bl. f. Bakt. 34, 4 (1903) Dunin and Borkowski, J., 205* Anz. d. Akad. d. W. Zsepsch-Krakan, No. 7B, 608 (1910) DupRE, Fr. and Hermann, 348* Durham and Gruber, 202-'^" DuRiG, A., 216, 289 Pfliiger's Arch. 85, 401-504 (1901) Ebbecke, W., 342* Biochem. Zeitschr. 12, 485 (1908) Eberstadt, O., 428* Diss. Heidelberg (1909) Ebler, 88 Edinger, L. and Mayer, 421 Effront, Jean, 182 Ehrlich, P., 31, 195*, 197, 201, 246, 251, 360, 361, 430 Ivlin, Jahrb. 6 (1897) Münchner med. Wochenschr. Nos. 33, 34 (1903) Deutsch, med. Wochenschr. 597 (1898) and Bechhold, H., 394*, 396, 404, 408 and Unna, P. G., 424 Einstein, A., 50, 54 Eisenberg, 200, 201, 307, 402, 435 and Landst(;in(>r and Volk, 201 and Okolska, 403 and Volk, 200, 200*, 203 Zeitschr. f. Hygiene 40, 155 (1902) ElSSLER, 135 and Pring.sheim, 135 Ekroth, Clarence V., 169''""' Elias, 208 Ellenger and Spiro, 299 Ellis, Risdale, 87 Elsberg, C. A., 261 Elsner, Fr., 175 Emslander, f., 179*3, 179*, 180, 180*', 181*2 1) Ivoll. Zeitschr. 5, 25 (1909) 2) Koll. Zeitschr. 6, 156 (1910) 3) Koll. Zeitschr. 7, 177 (1910) and Freundlich, 180* Zeitschr. f. physilval. Chemie 49, 322 (1904) Rieh, 179-^", 180* Engelmann, Th. W., 296 Engels, 218*, 220 Arch. f. exper. Pathol, u. Pharmakol. 51, 346-360 (1904) Epstein. A. A., 339 Errico, G. d', 136*, 330 and Jappelli, 296 and Savarese, 330 Etienne, G., 375* Rev. med. de l'Est 1 (Sept., 1907) Euler, H., 53, 182 Ewald, R. and Strassburg, 226 Faraday^, Michael, 75, 125 Feist, C. and Bobertag, O. and Fischer, H. W., 67* Fellner, 88 Field, C. N. and Teague, O., 205* Journ. of Exper. Med. 9, No. 1, pp. 86-92 FiLEHNE, W., 324* Berliner klin.Wochenschr. No. 3(1898) Arch, inter, de Pharmacodynamic F, No. 133 (1900) FiLippi, E., 371, 372 Lo sperimentale 62, 503-522 (1908) Arch, italicnnes de biologie 50, 175- 189 (1908); 51, 447-456 (1909) and Rodolico, 372 442 AUTHORS' INDEX FiNDLAY, A., 135*, 147* Koll. Zeitschr. 3, 169 (1908) FiNKELSTEIN, H., 323 Fischer, 17, 37 A., 418, 421, 422 and Berg, Waltlier, 418 Emil, 5, 133, 182 H. W., 216*1, 384*2^ 385 1) Beitr. z. Biol. d. Pflauzer (1910) 2) Biochem. Zeitschr. 27, 223-245 (1910) Bobertag, O. and Feist, C, 67* Ber. d. D. Chem. Ges. 3675 (1908) and Brieger, E., 309 and Jensen, P., 222*, 292 Biochem. Zeitschr. 20, 143-165 (1909) Martin H., 12, 33, 36, 36^", 67*, 68, 68^'^-, 70, 115, 160*, 175^", 224^"-, 225, 226*, 227, 228*2, 229'"-, 229, 230, 230*1, 231, 232, 233, 236, 267, 282, 289, 289*, 290, 306*^", 315, 316, 320*, 321, 321^", 323*, 323^''. 324, 327, 328^", 332^", 333*, 334, 336, 338, 338*1, 339, 344*i, 352. 410, 411 Das Oedem. in exper. v. Therapeut. Unterricht d. Physiol, u. Pathol d. Wasserb. im. Organismus (Dres- den, 1910) Die Nephritis, eine exper. u. kritische Studie über Natur u. Ursachen, d. Prinz, ihrer Bechandlung (Dres- den, 1911) 1) Kolloidchem. Beihefte. 2, 304. (1911) 2) Koll. Zeitschr. 8, 159-167 (1911) and Bayliss, 220^« and Clowes, 12, 175 and Henderson, L. J., 232^" and Hogan, 220^'^ and Hooker, M. O., 33, 36, 36^^ 352 and Jensen, P., 291* and Losev, G., 27* and Streitmann, A., 296 and Sykes, A., 334 and Woodyatt, 335 FiSKE, 188 Fleischmann, 208 Fletcher, W. M. and Hopkiijs, F. G., 298 and Langley, 328 Flxjri, M., 245* Flora 99, 81-126 (1908) FoA, C. and Aggazzotti, A., 374*, 384* Biochem. Zeitschr. 19 (1909) Fornet, A., 178 FouARDi, E., 102*, 107, 133*, 134, 134*, 180 KoU.-Zeitschr. 4, 185 (1909) L'Etat colloidal de l'Amidon (Laval, 1911) Fränkel, S. and Hamburg, M., 190* Beitrag z. chem. Physiol, u. Pathol. 8, 389-398 (1906) Frank, E. 235* Zeitschr. f. physiol. Chemie 70, 129- 142 (1910) Fränkel, S., 183 Frankl, 413 Arch. f. exper. Path. u. Pharm. 57 (1907) Frei, W., 311* Zeitschr. f. Infektionskrank, u. Hyg. d. Haustiere 6, 363-373, 446-475 (1909) Margadant, 396, 403 Freundlich, H., 14, 23, 23*, 23^"-, 24, 24^"-, 25, 25^", 28*3, 29, 66, 78"'-. 79, 80, 84, 87, 110, 147^", 187, 360,, 394*1 1) Habilitationsschrift (Leipzig, 1906) 2) KoU.-Zeitschr. 1, 321 (1907); 7, 193 (1910) (and Emslander), 180* and Losev, G., 27 Zeitschr. f. physikal. Chemie 59» 284-312 (1907) and Michaelis, L., 79 and Ostwald, Wo., 147-^" and Schucht, 25* (and Straub), 243 Frey, E., 334*, 338, 410* Pflüger's Arch. 120, 66-136 (1907) Friedberger, 209, 210, 371 and Isuneoka, 365 Friedemann, U., 149, 149*, 156*i, 157, 202-^", 207, 323*2 1) Arch. f. Hygiene 55, 361-389 (1906) AUTHORS' INDEX 443 Friedemann, TJ., -) in Oppenheimer's Handb. cl. Bicdiemie 3, 2 and Friedenthal, H., 149*, 160*, 202* Zeitschr. f . exper. Pathol, u. Therapie 3, 73-88 (1906) and Neisser, M., 84*, 86, 203, 203*, 205*, 283* Friedenthal, H., 6, 135*i, 149*, 157*, 164* Ber. d. D. Chem. Gcs. 44, 906 (1911) 1) Physiol. Zeutralbl. 12, 819 (1899) and Friedemann, U., 149*, 160*, 202* Friedländer, J., 64*, 136* Zeitschr. f. physikal. Chemie 38, 430 (1901) 2FÜRTH, O. VON, 169 and Schütz, J., 191* Beitr. z. chem. Physiol, u. Pathol. 9, 28-49 (1907) and Lenk, 291 Gabritschewsky, a., 286 •Gaidukow, N., 127, 277, 278 DmikeLfeldbeleuchtung u. Ultrami- kroskopie in d. Biologien. Medezin (Jena, 1910) Galecki and Zsigmondy and Wilke- Dörfürt, 98 Galitzer, S., 204 Galeotti, a., 340* Arch. f. Anat. u. Physiol. Abt. 200 (1902) Gartner, A., 172 -Gatin-Gruszewska, Z., 136* Pflüger's Arch. 103 (1904) Gay-Lussac, 51 and Boyle, 51 and Southard, 377 Gebhardt, W., 263, 269 Gengou, O., 371 and Bordet, 207 Gerhartz, H., 216*, 265, 265*i, 266 Pflüger's Arch. 133, 397-499 (1910) 1) Pflüger's Arch. 135, 104-170 (1910) Gerloff and Kihiitz, 263'^"" Gesell, R. A., 333 GiBBS, W., 25, 25^'' GiEMSA, G., 434 Gieson, van, 435 Gilbert and Lawes, 265 Girard, 59*, 322, 322* Comptes rend, de i'ac. d. Lc. 146, 927 (1908); 148, 1047-1186 (1909j; 150, 1446 (1910) Journ. d. physiol. and pathol. Gen. 12, 471 (1910) -Mangin and Henri, V., 204* Glauber, 8 Godlewski, 88* Goldschmidt, R. and Pribram, 387* Zeitschr. f . exper. Pathol, u. Therapie 6, 1 (1909) GoLGi, 264 GoLL, 332 Golodetz, L., 163 GoLODETZ, L. and Unna, P. G., 434 GoM, W., 366 GoocH, 98 Goppelsröder, 108 Gottlieb, R. and Magntjs, R., 332, 336* Arch. f. exper. Pathol, u. Pharm. 45, 223 (1901) and Meyer, H., 332*, 380*, 412* Graham, Thomas, 3*, 4, 10*, 55, 84, 90, 146 Philos. Transactions 1, 183 (1861) Liebig's Aim. d. Chemie 121 (1862) and Herzog, R. O., 146*6 and Stephan, 54 Gram, 30, 435 Grawitz, B., 418^", 421 Gros, O. and O'Connor, J. M., 365*, 372* Arch. f. exper. Pathol, u. Pharm. 64, 456-467 (1911) Gregor, 298 Pflüger's Arch., 101 (1904) Grosser, O., 102, 174, 350 Grüber and Durham, 202-'^" and Widal, 202^» Grübler, 186 Grünwald, 411 Grützner, P. von, 353 GUAGLIERIELLO, 143* GuERRiNi and Neisser, M., 288* Guggenheim, M. and Abderhalden, 34*, 189* Gully, Baumann, 329 GUMILEVSKY, G. O., 318* Pflüger's Arch. 39, 566 (1SS6) 444 AUTHORS' INDEX GÜRBER, A., 245, 307, 314 and Limbeck, V. and Hamburger, H. J., 321 Haak, a., 124^"-, 125 Haas, A. R. C, 388 Haber, f.; 59* Ann. d. Phys. (4) 26, 927 (1908) Zeitschr. f. physikal. Chemie 67, 385 (1909) HÄBERLE and Vorländer, 19-^" HÄCKEL, 252 Hahn and Trommsdorf, E,., 206 Münchner med. Wochenschr, No. 19 (1900) Hailer, E., 199* Arbb. d. k. Gesundheitssamts 29, H. 2 (1908) Hales, 66 Halliburton, 219* Hamburg, M. and Fraenkel, S., 190* Hamburger, H. J., 236, 244, 254, 307, 322, 322*2 2) Biochem. Zeitschr. 11, 443-480 (1908) and Hekma, 287* Biochem. Zeitschr. 3, 88-108; 7, 102- 116 (1906-1907); 9, 275-306, 512- 521 (1907); 24, 470-477 (1910) Koeppe, H. and Overton, E., 236 and Limbeck, von, 314 and Limbeck, von and Gürber, A., 321 Handovsky, H., 82*3, ng^ 147/«^ 149*1^ 151, 152*2, 155, 362*2 /« 1) Koll.-Zeitschr. 4 and 5 (1910) 2) Biochem. Zeitschr. 25, 510 (1910) and Pauh, Wo., 147^«, 149*i, 151, 311 Hannover, 421 Harden, A. and Young, W. J., 191 Proc. Roy. Soc. 77, B, 405-420 (1906); 78, B, 369-375 (1906) Hardy, W. B., 10, 64, 145, 158, 158*i, 159, 159*2 1) Journ. of Phys. 33, 251-337 Proc. Roy. Soc. 79, 413-426, Ser. B (London) Harlow, M. M. and Stieles, P. G., 189* Journ. of Biol. Chem. 6 (1909) Harnack, E., 142 Harriman, Mrs. Oliver, 169-^'' Harris and Donnan, 46 Harrison, W., 135* Harvey Lectures, 210, 228^", 299 Hatmaker and Just, 178 Hatschek, E., 15, 15*, 16, 16*, 262, 348 Koll.-Zeitschr. 6, 254-258 (1910); 7,. 81-86 (1910) Hansen and Drey er, 144 Hausmann, J., 262* Zeitschr. f. anorgan. Chemie 40, 110- 145 (1904) Haversian, 263 Hay, 413* Journ. of Anat. and Phys. 16 and 17 Heald, F. D., 382 Hecker, 156, 157, 206 and Pauh, 156, 157 Hedin, S. G., 28*4, 185, 191*2, 192*,, 243*1, 307 1) Pflüger's Arch. 60, 360 (1895) 2) Biochem. Journ. 1, 484-495 (1906)^ 3) Biochem. Journ. 1, 474, 484; 2,, 27, 81, 112 (1906) ") Zeitschr. f . physiol. Chemie 50, 497' (1907); 60, 364 (1909) Heidenhain, M., 318 Pflüger's Arch. 56, 579 (1894) Heinz, 408 Hekma, 160, 279 and Hamburger, H. J., 287* Helbig, 435 Held, 234, 234* Arch. f. exper. Pathol, u, Pharm. 53;. 227 Helmholtz, 353 Henderson, 310 Lawrence J. and Fischer, M. H.,. 232^" Henri, V., 83*, 187 Comptes rend. 147, 62-65 (and Girard-Mangin), 204* Henry (laws), 20, 242, 243, 308, 309, 388, 396 Herbst, C, 264 Mitterlungen d. zool. Station NeapeL 11, 185, 191 (1803) Herman, H., 293 Hermann and Dupre;, Fr., 348 Pflüger's Arch. 26, 442 AUTHORS' INDEX 445 Herzog, B. and Adler, 27* R. O., 53, 53^", 54, 103, 104, 146, 146*" Zeitschr. f. Elektrochcm. 13, 533- 539 a 907) and Arrhenius, Sv., 47 (and Graham, Th.), 146*« and Kasarnowski, 104*, 183 , 190* Biochem. Zeitschr. 11, 172 (1908) and Öholen, L. W., 53^" Hess, W., 114 Münchner med. Wochenschr., No. 32 (1907) Hessberg, P., 208, 209 Heyden, von, 99, 366, 376 Hill, A. Croft, 250 A. C. and Parnas, 298 HiROKOWA, 411 Hersch and Beck, 113 Hirschfeld, M. and PauU, 152 Hirshfeld, L., 204, 204*^, 285, 285*, 286 1) Zeitschr. f. allg. Phys. 9, 529- 534 2) Arch. f. Hyg. 63, 237-286 Höber, R., 81, 243*«, 244*^, 292*, 294*, 295, 295*«, 295*", 296, 298, 307, 307*^ 311, 311*2, 319*1^ 322, 329, 353, 353*3", 382, 386, 387, 387** 1) Pfluger's Arch. 70, 624 (1898); 75, 246 (1899) 2) Pfluger's Arch. 81, 522 (1900) •■>) Physikal. Chem. d. Zelle (Leipzig, 1906) 3") Physikal. Chem. d. Zelle u. d. Gewebe (Leipzig, 1911) *) Pfluger's Arch. 120, 508 (1907) 5) Physikal. Chem. u. Physiol, in A. V. Koränyi u. P. F. Richters Physikal. Chemie u. Medizin 1, 389 (Leipzig, 1907) «) Biochem. Zeitschr. 14, 209 (1908) ") Biochem. Zeitschr. 14, 209 (1908) Pfluger's Arch. 134, 311 (1910) 8) Biochem. Zeitschr. 17, 518 (1909) 9) Zeitschr. f. Physikal. Chemie 70, 134-145 (1910) 10) Zeitschr. f. aUg. Physiol. 10, 173- 189 (1910) HöBKR, R., ") Arch. .^ d. ges. Physiol. 134, 311-336 (1910) and Chassin, S., 427* KoU. Zeitschr. 3, 76 (1908) and Joel, A., 386 and Ruhland, 242 and Waldenberg, 295* Pfluger's Arch. 126, 331 (1909) HoFF, J. H. van't, 43 Hofmeister, F., 67*, 95, 147^", 151, 166, 220, 276, 277, 318, 319, 337, 343, 409 Arch. f. e.xper. Pathol, u. Pharm 27, 395 (1890^; 28, 210 (1891) and Ostwald, Wo., 337 and Pick, 166 P., 115 HoGAN, J. J. and Fischer, W. H., 220''' Holde, D., 11*, 64* Chemiker. Ztg., No. 54 (1908); Zeitsciir. :'. angew. Chemie, 2138ff, (1908) Holderer, 187 Holer, R., 230 hollinger, 235* Biochem. Zeitschr. 17, 1 (1908) Hooker, D. R., 231, 332 Am. Journ. of Physiol. 27, 24-44 (1910) Marian O. and Fischer, M. H., 33, 36, 36^", 352 Hopkins, F. G. and Fletcher, 298 Hoppe and Seyler, 226 Howell, 299 HowLAND and Marriot, 314 Hunt, R., 366 Illyes, G. and Kövesi, 340 Inada, R. and Müller, O., 311, 380 IscovESCO, H., 185*^ 239, 299, 300*^, 330*2, 342*2^ 375/" 1) Comptes rend, de la Soc. de Biol. 60 (1906) 2) Etude sur les humeurs de I'organ- ique (Paris, 1906) 3) Biochem. Zeitschr. 24, 53-78 (1910) Ishizaka and Schucht, 84*^ Isuneoka and Friedberger, 365 446 AUTHORS' INDEX IzAR, G., 366*^ 372*1, 375*2^ 375^ 370*2 1) Biochem. Zeitschr. 20, 249, 266 (1909) 2) Zeitschr. f. Immunitätsforschun- gen, Oris?. 2 3) Zeitschr. f. klin. Medezin 68, 516 and Ascoli, M., 211, 365% 366, 371, 373*, 377*, 383^" Jacobson, P., 187 Jacoby, M., 184*, 196 Biochem. Zeitschr. 4, 21 (1907) and Schütze, A., 189*, 196 Zeitschr. f. Immunitätsforschungen 4, 730-739 (1910) Jacque, L. and Zunz, E., 199* Arch, intern de Physiol. 8, 227-270 (1909) Jagic, N. and Landsteiner, K., 200*, 204* Japelli, G. and D'Errico, 296 Jensen, P. and Fischer, H. W., 212, 222* and Fischer, M. H., 291* Joel and Höber, R., 386 Johnson, 184 Joos, J., 201* Zeitschr. f . Hygiene 40, 203 (1902) JoRDis, E., 93, 93^"- Joseph, 377 Dermat. Zentrabl. (Sept. 1907) JOSLIN, 314 JosT, L., 259* Tharandter forst. Jahrbuch 60, 331- 334 (1909) JUNGFLBISCH, 20 JÜNGSEN and (Sörensen), 143* Just and Hatmaker, 175 Kahlenberg and True, 382 Kasarnowski, H. and Herzog, R. O., 104*, 183*, 190* Katz, J. R., 178, 292 Katzenellenbogen, 319* Pflüger's Arch. 114, 522 (1906) Kaufmann, M., 366 Keesom, 119 KiLNiTZ and Gerloff, 263*''* Kirschbaum, 99, 102, 197, 205 and Pribram, 99 Kisch, 240 Klatte and Büchner, 184* Klemensiewicz, 231 Klemperer, G., 343, 343-^" Klose and Vogt, 352 Kneipp, Fr., 364 Knoevenagel, E., 70, 187, 428 Sitzung d. ehem. ges. zu Heidelberii' 17, 2 (1911); (Zeitsclir. f. ang. Chan. 505 (1911) KoBER, P. A., 120, 120^"- Kobler, 173 Koch, Robert, 138, 193, 395, 396 Koeppe, H., 307 and Hamburger and Overton, 236 Köhler, F. and Traube, J., 163, 427/» KOHLER, R., 95 Kolle, von, 196-'^" KoPACZEWSKi, 94, 94*"- KoRANYi, A. VON, 315, 339*, 341 Zeitschr. f. Idin. Medezin 34 (1898) BerUner Idin. Wochenschr. 781(1899) and Bence, J., 315 and Kövesi, 341 Kövesi and Roth-Schultz, 340* Pathol, u. Therapie d. Merenin- suffizienz 76 (Leipzig, 1904) and Richter, O., 315 Physikal. Chemie u. Medizin (Leip- zig, 1907-1908) K0RÄNYI, P. T. and Richter, A. v., 113 Kossel, a., 160, 279 Münchner med. Wochenschr. 58, 2 (1911) Kövesi and Illyes, G., 340 A. V. and Koränyi, A. von and Roth- Schultz, 340* andSuranyi, 341 Krafft, f., 46 and Smits, A., 45, 46 Krans, 313 Kreidl, a. and Lenk, 350* and Neumann, 349* Kröenig, 50 Krönig and Paul, 395, 401, 402, 403 Kubowitsch, J. A., 266 Kunde, 216 Kundt, 76 Kupfer, von, 248, 373 KtJsTER, E., 264 AUTHORS' INDEX 447 Laer, W. H. van, 180, 184 T. Congres intern, de Brasserie; 25, 7 (1910) Bull. Acad. R. Belg. 305-320; 302- 370 (1911) Lagergreen, 337 Lamy, E. and Mayer, A., 334*, 335 Comptes rend, de la Soc. de Biol. 222 (1904) Landstetner, K., 196^", 198, 200, 201, 205 and Botteri, A., 199* Zentralbl. f.Bakt. 42, 562-5G6 (190G) Eisenberg and Volk, 201 and Jagic, N., 200*, 204* Münchner med. Wochenschr., No. 27 (1904) and Pauli, Wo., 205*, 205 Wiener med. Wochenschr., No. 18 (1908) and Uhlirz, 145* Centralbl. f. Bakter. 40, 206 (1905) and Weleck, St., 200 Zeitschr. f. Immunitätsfor. u. exper. Therapie 8, 397-403 (1910) XiANDWEHR and Baisch, 343 Langley, J. N. and Fletcher, 328 Laqueur, E., 154, 164 and Sackur, O., 154*, 164, 164* Beitr. z. chem. Physiol u. Pathol. 3, 193, 224 Lawes and Gilbert, 265 Lea, Carey, 366 Am. Journ. of Science (3), 37, 476; 38, 47; 41, 482 Leavenworth and Mendel, 266 Lebedew, a. von, 102*, 190* Biochem. Zeitschr. 20, 114-125 (1909) Leber, Th., 286 Leclef and Denys, 288 Leduc, Stephane, 254, 254^", 255, 255^"-, 256, 256^^'-, 257 Lemanissier, J., 73*, 144, 144*, 349* L'Etude des corps ultramicrosc. Thesis (Paris, 1905) L'Etude des corps ultramicrosc, Rousset (Paris, 1905) and Michaelis, 144 Lenk, E., 169, 350* and Fürth, O. von, 291 andKreicU,A.,350* Lennep, Ross van, 181 Lepeschikin, W. W., 240, 245, 278 Levaditi and Yamanouchi, 208 Levites, S. J., 162 Lewith, 151 Lichtwitz, L., 342*^, 343, 363, 364 1) D. med. Wochenschr., No. 15 (1910) -) Zeitschr. f. physiol. Chemie 64, 144-157 (1910) and Rosenbach, F. J., 342*, 343* Zeitschr. f. physiol. Chemie 61, 117 (1909) Lieber and Schmidt, P., 189 Liebermann, L., 152*, 153*, 177 L. von, 206, 265 Liesegang, R. E., 106*i, 239, 257, 260*3, 261, 261*1, 261'"-, 263, 263*% 264, 269*, 269*^ 1) Chem. Reaktionen in Gallerten (Leipzig, 1898) 2) Beitr. z. einer koUoidchem. d. Lebens (Dresden, 1909) 3) Ann. d. Phys. 19, 406 (1906); 32, 1095 (1910) 4) Koll. Zeitschr. 7, 219 (1910) 5) Naturw. Wochenschr., No. 41 (1910) ») Journ. f. Phychol. u. Neurol. 17, 1-18 (1910) LiLLiE, K. G., 378 Am. Journ. of Physiol. 17, 89 (1906); 24, 459 (1909) R. S., 295*2, 354 Am. Journ. of Physiol. 10, 419 (1904) Limbeck, V., Gürber, A. and Ham- burger, H. J., 321 von and Hamburger, 314 LiPMAN, C. B., 379 Loeb, Jacques, 225^", 230, 241, 245, 245*^ 264*3, 282, 289*, 378, 379, 413*2 1) Pflüger's Arch. 69, 1 (1898); 71, 457 (1898); 75, 303 (1899) 2) Pflüger's Arch. 91, 248^(1902) 3) Untersuch, üb. d. künst. Parthe- nogenese (Leipzig, 1906) «) Biochem. Zeitschr. 11, 144 (1908) J. and Bentner, R., 240, 295, 295^" and Osterhout, W. J. V., 241 448 AUTHORS' INDEX LÖPPLER, 30 LoEPER, 321*, 412* Comptes rend, de la Soc. de Biol. 58, 1056 (1905) LoEWE, G., 243 S., 139, 386 LoEWY, A., 308*2, 309, 312*i 1) in A. von Koränyi u. P. F. Rich- ters Pathol, d. Respiration 2, 37 (Leipzig, 1908) 2) in A. von Koränyi u. P. F. Rich- ters Physikal. Chemie u. Medezin 1, 248 (Leipzig, 1908) and Munzer, E., 313* Arch. f. Physiol. (1901) LÖFFLER, 30, 30-^" Lorenz, R., 11, 49^", 51 LosEv, G. and Freundlich, H., 27* Lottermoser, A., 28*, 84*, -365^" Koll. Zeitschr. 6, 78-83 (1910) Lowe, Chas., 252^" Löwe, H., 362 lubarsch, 231 LuDERKiNG, Chas. and Wiedermann, E., 134*, 137* Ludwig, C, 66, 336, 337 Ludwig, G., 33r" Luther (-Ostwald), 108^", 113 Macallum, a. B., 26, 234, 234-^", 287, 292, 296, 296*, 298 Science (Oct. 7, 14, 1910) MacCallum, 413, 413* 1) Pflüger's Arch. 104; Am. Journ. of Physiol. 10, 101, 259 2) On the mechanism of the physiol. action of the cathartics (Univ. of Cal. pubUc, 1906) MacDonald, 298A, 354 McClendon, J. M., 98, 379, 405 MCCOLLUM, 351 Madsbn, Th., 56, 105 Magnus, R., 190, 190*, 227, 333*i 1) Arch. f. exper. Pathol. 45, 210 (1901) ^) Zeitschr. f. physiol. Chemie 41, 149-154 (1904) Magnus, R. and Gottlieb, R., 332, 336* Mai, J. and Rothenpusser, S., 174 1) Milchwirfcschaftl. Zeutralbl., No. 4 (1910) 2) Zeitschr. f. Unters, d. Nahrungs- genussmittel, No. 12 (1909) Maier and Forges, O., 208 Malfitano, G., 91, 93^'^-, 95, 102 and Duclaux, J., 92, 102 and Moschkoff, A. N., 134* Comptes rend, de I'acad. d. sciences 151, 817ff. Mangin and Henri and Girard, 204 Mansfeld, G., 388* Pflüger's Arch. 131, 457-464 (1910) and Bosanyi, 386 Marc, 120, 121, 180 Rob, 26* Marchand, 23 Margadant and Frei, 396, 403 Marinesco, 266 Marriot and Howland, 314 Martin, C. J., 143*, 146* Masius, M. and Michaelis, L., 337 Massard and Bordet, J., 286 Masuda, 353 Mathews, 353, 353*, 382, 382* Science 15, 492 (1902) Matruchot, p. and Molliard, 216* Rev. Zen. de Botanique 14 (1902) Matuta, j., 152, 156-^" Mauabe, 152 Mayer, Andre, 159, 159* Comptes rend. (Oct. 8, 1906) and Edinger, 421 A. and Lamy, E., 334*, 335 and Schaeffer, G., 279* Comptes rend, de la Soc. de Biol. 64, 681 (1908) Mayerhofer, E. and Pribram, E., 322* 1) Wiener khn.Wochenschr. 25 (1909) 2) Zeitschr. f. exper. Pathol, u. Thera- pie 7 (1909) 3) Biochem. Zeitschrift 24, 453-469 (1910); 27, 376-384 (1910) Mechowski, W., 363 Mecklenburg, W., 25, 25*, 119, 120 Meigs, E. B., 290, 290*, 291* Am. Journ. of Physiol. 26, 191-211 (1910) AUTHORS' INDEX 449 Meijeringh, W., 175* Chem. Weekblivl 7, 951-953 Meltzer, S. J., 386, 387 and Shaklee, A. O., 189* Mendel, L. B. and Leavenworth, 266 Merck, 99 Merino, von, 323 Mbstrezat, 354 Metcalf, W. v., 33*, 34, 35, 347* Zeitschr. f. physikal. Chemie 52, 1 (1905) (and Ramsden), 348 Metchnikoff, Elie, 285 E., 193, 239 Meurer, R., 245* Jahrb. f. wessensch. Botanik 46, 503- 567 (1909) Meyer, H. and Gottlieb, R , 332*, 380*, 412* Exper. Pharmakologie (Berlin, 1910) Hans and Overton, E., 385, 386, 387, 388 Kurt, 54* Hofmeisters Beitr. z. chem. Physiol, u. Pathol. 7, 393Ef. Michaelis, L., 27, 28, 77, 107, 113, 119, 119'"-, 141, 144, 144*1, 145^ 147, 147*3, i5§^ 161^ 164^ 165^ 185*2, 186, 187, 201, 202, 285 1) D. med. Wochenschr. 42 (1904); Virchows Arch. 179, 195-208 (1905) 2) Biochem. Zeitschr. 7, 488-492; 12, 26; 16, 81-86, 486-488; 17, 231- 234; 19, 181-185 (1907-1909) 3) Biochem. Zeitsclu". 19, 181 (1909) and Beniasch, 204 and Davidsohn, 202 and FremidUch, 79 and Lemanissier, J., 144 and Masius, 337 andRona, P., 107, 107*^ 141*i, 145*, 147, 201, 234, 348* 1) Biochem. Zeitschi". 4, 11 (1907) 2) Biochem. Zeitschr. 14, 476ff (1908) ») Biochem. Zeitschr. 15, 196 (1908) ■*) Biochem. Zeitschr. 21, 114-122 5) Biochem. Zeitschr. 25, 259-^36() (1910) and Skwirsky, 206 MicuLiciCH, M., 307* Zcntralbl. f. Physiol. 24, 12 Minkowski, O., 314 MiTTELBACH, Robert, 92-'^" Modelski, J. W. and Pfeiffer, 156 MoLiscH, H., 216, 217^"- Moll, 159* Hofmeisters Beitr. 4, 563 (1903) MoLLiARD and Matruchot, P., 216* Moore, A. R., 230 and Roaf, H. E.. 387* B. G. and Roaf, H. E., 73* Proc. Roy. Soc. of London, Ser. B. 73, 382 (1904); 77, 86 (1906) MoRAWiTz, P., 394* Kolloidchem. Beihefte 1, 301 (1910) Morgan, J. L. R , 109 Morgenroth, J., 201*i, 205*^ 1) Münchner med. Wochenschr., No. 2 (1903) 2) Arbb. a. d. Pathol. Inst. (Festschr.) (Berlin, 1906) and Pane, D., 206*^, 206* Biochem. Zeitschr. 1, 354-366 (1906) Mörner, K. a. H., 72*, 343*2 1) Zeitschr. f. phj'-siol. Chemie 34, 207 (1901) 2) Skandinav. Arch, f . Physiol. (1901) MoROCHOWETz, Leo, 94, 94*"- Morse, H. W. and Pierce, G. W., 262* Zeitschr. f. physikal. Chemie 45, 589-607 (1903) MoRUZzi, G., 311* Biochem. Zeitschr. 28, 97-105 (1910) MoscHKOFF, A. N. and Malfitano, 134* Motzfeld, 339* MOUFANG, E., 180 Much, H., 144, 196 Römer and Siebert, C, 144* Zeitsclir. f . cüätet. u. physikal. Thera- pie 8 (1904, 1905) Mühlmann, 266 MÜLLER, O. and Inada, R., 311, 380* Deutsch, med. Wochenschi-. (1904) and Thürgan, H., 216, 256*, 257^"' Zentralbl. f. Bakter. 20 (II) No. 12/14; 51/17 (1908) Munch, 296 450 AUTHORS' INDEX MuNK, F., 208 M., 264 MÜNTZ, A., 215* Comptes rend, de I'acad. d. sciences 150, 1390-1395 (1910); 151, 790- 793 (1910) MuNZER, E. and Loewy, A., 313* MUTH, W., 112* Nagel, G., 33* Ann. d. Physik. (4) 29, 1029-1056 (1909) Nägeli, a. E., 215, 216 O., 278 Nailashima, 247 Nathan, E. and Sachs, H., 210 Nathanson, 240 Neisser, M., 86, 202*, 435 Hys;ien. Rundschau 13, 1261 (1Ö03) and Friedemann, U., 84*, 86, 203, 203*, 205*, 283* Mi'mchner med. Wochenschr., No. 11, 19 (1904) and Guerrini, 288* Arbb. a. d. Kgl. Inst. f. exper. Thera- pie, No. 4 (1908) and Sachs, A., 207 N'ell, Peter, 54* Drude's Ann. 18, 323 (1905) Nernst, W., 20, 63, 122 Neubauer, E. (and Porges, O.), 87*, 140*, 141*, 388* O., 208 Neuberg, C., 381, 381* 1) Sitzg. d. D. Chem. Ges. v. VI, 7 (1904) 2) Biochem. Zeitschr. 1, 166 (1906) 3) Koll. Zeitschr. 2, 321, 354 (1908) and Albu, A., 218*, 219*, 233* Neufeld, 288 Neumann, A., 247* Zeutralbl. f. Physiol. 21, 102-105 and Kreidl, A., 349* NiCLOux, M., 388* Les anesthesiques generaux (Paris, 1908) NoFF, 299 NovY, J., 328 and Vaughn, 210 Obermeyer, Fr. and Pick, E. P., 197* Wienerkhn. Wochenschr. (1906) O'Connor and Gros, O., 365*, 372* Oden, S., 80 and Ohion, Sl^"" and Pauli, Wo., 152 Oesper, 17^" Ohlon and Oden, 81-^" ÖHOLM, L. W., 53, 53^", 54, 103*, 103*"" Zeitschr. f . physikal. Chemie 50, 309- 349 (1904) ; 70, 278-407 (1909.^ and Herzog, R. O., 53^" Qkbr-Blom, M., 307, 319* Skandinav. Arch. f. Physiol. 20, 102- 114 (1907) Okolska and Eisenberg, 403 Omorvkow and Sachs, H. and Ritz^ 196 Onnes, Kamerlingh, 119 Oppenheimer, C, 182 Kurt, 174 OsTERHOUT, W. J. v., 241, 241^«, 379,. 386 1) Botanical Gazette 46, 53-55 (1908) 2) Journ. of Biol. Chem. I, 363-369- (1906) and Loeb, J., 241 OsTWALD, Walter and Riedel, A., 179 Wilhelm, 5^", 28*, 64, 113, 178, 259, 262*2 1) Zeitschr. f. physikal. Chemie, 62^ 512 (1908) 2) Lehrt, d. allgem. Chemie 2, 11 (2"<^ ed.) Wolfgang, 12, 13, 17^", 26, 66, 67*\ 70, 71^'^, 87, 147^", 161, 309, 309*3, 310, 379, 379*1, 330, sgO** 1) Pflüger's Arch. 106, 568 (1905) 2) Pflüger's Arch. 108, 563 (1905) 3) Koll. Zeitschr. 2, 264, 294 (1908) 4) Koll. Zeitschr. 6, 297 (1910) and Freundlich, 147-^" and Hofmeister, 337 Luther, 108^'^, 113 Sprengel, 113 Oswald, A., 232*, 233 Zeitschr. f . exper_Pathol. u. Therapie 8, 2ß6 (1910) Ottenberg, R., 190 Ottolenghi, 408 Desinfektion 2, 109 AUTHORS' INDEX 451 Overton, C. E., 66*, 236, 239, 240, 241, 242, 289*, 294, 294*, 353 1) Pflüger's Arch. 92, 115 (1902) 2) Pfliiger's Arch. 105, 176 (1904) 3) Biochem. Zeutralbl. 2, 518 «) Verh. d. Ges. D. Naturf. II, 416 (1903) and Hamburger and Koeppe, 236 and Meyer, Hans, 385, 386, 387, 388 Paal, C, 31, 86, 99 Padtberg, 234 Pane, D. and Morgenroth, J., 206*, 206*5 Paneth, 88 Park, 359 Parnos and Hill, A. V., 298 Parodko, Th., 282 Pasteur, L., 193 Patin, G. and Roblin, L., 373* Journ. Pharm, et Chemie (6) 30, • 481-483 Paul, H., 222* Mittlgn. d. k. Bayr. Moorkulturan- stalt No. 2 (1908) Th., 366 and Krönig, 395, 401, 402, 403 Pauii, Wo., 67*S 68, 82, 86, 113, 115, 137*S 147, 147^", 148, 149*S 151, 152, 154, 156, 156*S 157, 205, 293, 343, 381**, 410*3, 413*3 1) (Pascheles) Pflüger's Arch. 67, 225 (1897) 2) Pflüger's Arch. 67, 219 (1897); 71, 1 (1898) 3) Verh. d. Kong. f. inn. Med. 21, 396 (1904) Sitzungber. d. k. Akad. d. Wiss. 113 (1904); 115 (1906) 4) Verh. d. 21 Kong. f. inn. Med. 5) KoU. Zeitsclir., No. 5, 1, 241 (1910) and Handovsky, H., 147^", 149*i, 151, 311 1) Beitr. z. chem. Physiol, u. Pathol. 9, 419 2) Biochem. Zeitschr. 18, 340ff . (1909) 3) Biochem. Zeitschr. 24, 239-262 (1910) and Hecker, 156, 157 and Hirschfeld, M., 152 Pauli and Landsteiner, 205, 205* and Oden, S., 152 and Rona, 162, 162* Hofmeisters Beitr. z. chem. Physiol, u. Pathol. 2, Iff. andSamec, 148*, 154, 161*, 268, 268* Biochem. Zeitschr. 17, 235-256 (1909) Pekelharning, C. a., 286 Pelet and Jolivet, L., 427* Die Theorie d. Färbesprozesses (Dresden, 1910) Pemsel (and Spiro), 152* Perkin, Sir Wm. Henry, 252-^" Perrin, J., 9, 51, 52"'-, 78, 79 Pettenkofer, 352 Pettibone and Abderhalden, 185* Pfaundler, M., 268 Pfeffer, W., 57*i, 221*=, 221^", 238*2, 286 1) Osmot. Untersuchungen (Leipzig, 1888) 2) Pflauzenphysiol. (Leipzig, 1897) Pfeffer, W. and Vries, H. dE, 239 Pfeiffer, P. and Modelski, J. W., 156 and Wittka, 156 Pflüger, 341 Philippi, E., 365* Philippson and Demoor, J., 298* Pick, A., 166, 167^", 362^" and Auerbach, 330 and Hofmeister, 166 E. P. and Obermayer, Fr., 197* • Pickering and Plateau and Quincke, 36-''" Pierce, G. W. and Morse, H. W., 262* PiNcussoHN, L., 144*, 230, 384 Biochem. Zeitschr. 10, 356 (1908) Plateau, Quincke and Pickering, 36-^" and Poggendorff, 33 Poggendorff and Plateau, 33 PoNFiCK, E., 333 ponomarew, 279 Porges, O., 87, 140, 141, 196^", 203, 208 and Maier, 208 and Neubauer, E., 87*, 140*, 141*, 388* 1) Biochem. Zeitschi-. 7, 152-177 (1907) 2) Koll. Zeitsclir. 5, 4 (1909) 452 AUTHORS' INDEX Portig, P., 365* Dissertation (Leipzig, 1909) POSNYAK, E., 116, 175 POTTEVIN, 250 Prescott, Samuel, 182 Preti, 365* Comptes rend. d. la Soc. de Biol. 65, 52, 224; Biochem. Zeitschr. (1909); Zeitschr. f. physiol. Chemie 58, 539; 60, 317 Pribram, B. E., 99, 102, 191*, 219*i, 221*1, 222*2, 296, 327*^ 1) KoUoidchem. Beihefte 2, 1 (1910) 2) Wiener klin. Wochenschr. 15 (1911) and Goldschmidt, 387* and Kirschbaum, 99 and Mayerhofer, 322* and Stein, E., 192 Pringsheim, H., 57*, 105*, 135*, 260 Jahrb. f. w. Botanik, 28, 1-38 (1895) and Eissler, 135 and Stoffel, 105* Quagliariello, G., 412* Biochem. Zeitschr. 27, 516-530 (1910) Quincke, G., 10, 18, 25, 35*2, sg/»^ 76*3, 78^ 108^», 137, 161, 300, 347, 347*2, 354 ^) Poggendorff's Annalen 139, Iff. (1870) 2) Wiedem. Ann. 35, 590 3) Ann. d. Phys. 1 (Appendix IV), 85-86 (1902) and Plateau and Pickering, 36-''" Rabl, H., 264 Naturforschervers. München (cit. by Liesegang) (1899) RÄHLMANN, E., 73*1, 127^ 136*1, 144, 144*2 1) Berhner kUn. Wochensch., No. 8 (1904) 2) Münchner med. Wochensch. 48 (1903) 3) Arch. f. d. ges. Physiol. 112, 128- 171 (1906) Rakowski, a., 133* Ramsay, Sir Wm., 54 Ramsden, W., 34*, 347* Arch. f. Anat. u. Physiol. (Physiol. Abt.) 517 (1894) Zeitschr. f. physikal. Chemie 47, 336 (1904) and Metcalf, 348 Ranvier, 287 Reichardt, 230, 231 Reichel, 395* Biochem. Zeitschr. 22, 149, 177, 201 (1909) Reichenbach, H., 404, 408, 408-''" Reicher, K., 247* Deut. med. Wochensch. 34 (2), 1529 (1908) Reid, E. W., 47 Reinders, E., 36, 238 Kon. Akad. van Wetenschappen Amsterdam. Proc. 563-573 (1910) Reinhold, B. and Riesenfeld, E. H., 80 Reinke, J., 66*, 116, 116*, 116'""- Hansteinsbotan. Abhand. 4, 1 (1879) Rettger, 299 Revello-Alves and Benedicenti, 157 Rhumbler, L., 283-^", 284, 285 Richter, A. von and Koranyi, P. T., 113 O. and Koranyi, 315 Richter, B. F. and Roth, W., 341 Riecke, E., 42 Riedel, A., and Ostwald, Walter, 179 Riesenfeld, E. H. and Reinhold, B., 80* Zeitschr. f. physikal. Chemie 66, 672-686 (1909) RiESMAN, 228 Ringer, W. E., 152^'^ and Adler, 379 Ringers, 137 RiTz, 189, 196 and Sachs, H. and Omarvkow, 196 RoAF, H. E. and Moore, A, R., 387* and Moore, B. G., 73* Robertson, T. B., 64*, 164, 164*, 166, 377 1) Journ. of Physic. Chem. 11, 542 (1907); 12, 473 (1908) 2) Koll. Zeitschr. 7, 7-10 (1910) 3) Die Physikal. Chemie d. Proteine (Dresden, 1911) AUTHORS' INDEX 453 Robin, A. and Weill, E., 371 RoBLiN, L. and Patin, G., 373* RoDEWALD, H., 6G, 134, 134*1 Zeitschr. f . physikal. Chemie 24, 193 (1897) RoDOLico and Filippi, 372 Roentgen, 18 ROGEE, 220^'^ Rogers, Allan, 169-''" Rogoff, J. M., 300 RoHDE, E., 33* Ann. d. Phys. (4) 19, 935 (1906) RoHLOFF and Schinja, 162 ROHONGI, 184 romanowsky, 434 Romberg, E , 380 Römer and Siebert, C. and Much 144* RoNA, P., 107*2, 108, 115*, 141*1, 145*^ 147, 201 and Michaelis, L , 348* and Michaelis, 107, 107*^, 141*i, 145*, 147, 201, 234 and PauH, 162, 162* Biochem. Zeitschr. 21, 114-122 (1909) and Takahashi, D., 235* 1) Biochem. Zeitschr. 30, 99-109 (1910) 2) Biochem. Zeitschr. 31, 33G-344 (1911) RoNDONi, P., 206, 206* Zeitschr. f. Immun, u. expcr. The- rapie 7, 515-543 (1910) and Sachs, H., 208 Röntgen, W. K., 81, 144 RoozEBOM, H. W. Bakhuis, 6 Rosenbach, F. J. and Lichtwitz, L., 342*, 343* ROSENTHALER, L., 183*, 188 Biochem. Zeitschr. 26, 9 (1910) ROSHARDT, P. A., 238* Beitr. z. Botan. Zentralbl. 25, Abt. I, 243-357 (1910) Roth, W. and Richter, P. F., 341 and Strauss, H., 323* Zeitschr. f. khn. Medizin 37, No. | RoTHE, A., 28 Rothenfusser, S., 174, 174*, 175, 181* Zeitschr. f. Unters, d. Nahrungs u. Genussm. 18, 135-155 (1909); 19, 261-268, 465-475 (1910) and Mai, 174* Rothmund, V., 64* RoTH-ScHULTZ and Koränyi and Kovesi, 340* Roux, W., 110, 256-''" and Yersin, 110 198* Ann. de I'lnst. Pasteur 3, 273-288 (1889) Rubner, 265 RuHLAND, 241, 242, 427 and Höber, 242 Rumpf, 219* Münchner med. Wochenschr., No. 9 (1905) Runge, E. F. (F. E.), 252, 253 RuNNSTRÖM, J. and Backmann, L., 265, 265* Russell, H. L. and Babcock, 175* Rysselberghe, B. VAN, 243* Mcm. de l'Acad. roy. de Belg. 58 (1899) Sabbatani, L., 363, 376 Sachs, H., 196, 198 and Altmann, 206, 208 and Neisser, M., 207 Omorvkow and Ritz, 196 and Nathan, E., 210 and Rondoni, P., 208 W., 216 Sackur, 164 O. and Laqueur, 154*, 164, 164* Salkowski, 343 Berliner klin. Wochenschr., No. 51, 52 (1905) Sansum, 143, 383 Santesson, 298* Skandinav. Arch. Phys. 14, 1 (1903) Salomon, 208 Samec, M., 133, 135*, 148*, 154, 156 and PauU, 148*, 154, 161*, 268, 268* Sasaki, Kumoji, 342* Hofmeisters Beiträge 9, 386 (1907) Savare, M., 342* Hofmeisters Beiträge 9, 401; 11, 71 Savarese and d'Errico, 330 Schade, H., 231, 266, 343, 344* 1) Kell. Zeitschr. 4, 175-261 (1909) 2) Kuli. ehem. Beihefte 1, 375 (1910) ^) Münchner med. Wochenschr., No. 1 2 (1909) 454 AUTHORS' INDEX Schade, H., *) Zeitschr. f. exper. Path. ■ u. Therapi 8, 2-34 ScHAEPFER, G. and Mayer, A., 279 SCHANZ, 144 SCHEITLXN, W., 310*, 311 Dissertation (Zurich, 1909) SCHELLENS, W., 400 Inaug. -Dissert. (Strassburg, 1905) ScHEURLEN and Spiro, 395, 401, 402 ScHiNYA and Rohloff, 162 Schleicher and Schüll, 92, 94, 95-''", 96 Schleicher, 95, 96 Schmidt, C. G., 26* P., 196, 205^" and Lieber, 189 Nielsen, Signe and Sigval, 34*, 189* Schneider, J., 81 Schoep, a., 95* Bull, de la soc. chem. de Belg. 24, 10 (1910) Schoep, A., 92^"- Schönborn, S. 329 Schorr, K., 152, 154, 166 Schroeder, p. von, 162, 361 ScHUCHT and Freundlich, 25* and Ishizaka, 84*^ ScHÜLL, 95, 96 and Schleicher, 92, 94, 95^", 96 Schulz, Fr. N. and Zsigmondy, E,., 143 ScHUMBERG and Zuntz, N., 216 Schütz, J. and Fürth, H. von, 191* Schütze, H. and Jacoby, M., 196 Schwarz, C., 291*, 294* Pfiüger's Arch. 117, 161 (1907) SCHWENKENBECHER, A., 324*, 345 Arch. f. Anat. u. Physiol. 121 (1904) Seddig, 50 Seyler and Hoppe, 226 SHAKLEE,A.O.andMELTZER, S.J.,189* Am. Journ. of Physiol. 25 (1909) Sherman, H. G., 169^" Sholto, J. and Drey er, 28* and Douglas and Dreyer, 200* SlEBECK, R., 336 SlEBERT, C., 144*, 196 and Römer and Much, 144* SiEDENTOPF, H., 76*1, 122, 123"'- 126^"-, 127 KoU. Zeitschr. 6, No. 1 (1910) and Zsigmondy, R., 6, 75 Signe and Schmidt-Nielsen, Sigval,. 34*, 189* Sjöquist, 152* Skwirsky and Michaelis, 206 Smoltjchowski, M. von, 50, 51 Smits, a. and Krafft, F., 45, 46 sobieranski, 411 Söhngen, 181 Sellman, Torald, 337 Sörensen and Jürgsen, 143* Southard and Gay, 377 Spiro, K., 67* Hofmeisters Beitr. z. chem. PhysioL 5, 276 (1904) and Bruno, J., 401*, 403* Arch. f. exper. Pharmak., 41 and Ellenger, 299 and Pemsel, 152* Zeitschr. f. physiol. Chemie 26, 233. (1898) and Scheurlen, 395, 401, 402 Sprengel (-Ostwald), 113 Stahl, E., 287 and de Bary, 286 Starling, E. H., 47, 332, 335,. 336* Steche and Waentig, 184 Stefan, 146 Stephan and Graham, 54 Stein, E. and Pribram, 192 Steiner, H., 28 Stern and Battelli, 386 Stiles, P. G. and Harlow, 189* Stodel, G., 365*, 371, 371^" Les Coll. en Biol, et en Ther. These- (Paris, 1908) Stoeltzner, W. and Dekhuysen, C.,. 420 Stoffel, F., 55*, 74, 172*, 262 Inaug.-Diss. (Zürich, 1908) G., 361*, 362* and Pringsheim, 105* Stöhr, 291^"- Strassburg and Ewald, R., 226 Strassburger, E., 238 Straub, W., 360*, 361* Pfiüger's Arch. 98, 5/6 and Freundlich, 243 Strauch, F. W. and Abderhalden^ 187 AUTHORS' INDEX 455 Strauss, E., 15ö H., 323*1, 329, 329*=, 339, 341, 345 1) Zeitschr. f. klin. Med. 57, No. f 2) in Koränyi and Richter 2, 110 and Roth, W., 323* Streitmann and Fischer, M. H., 296 Stumpf, 364 SvEDBERG, Th., 4, 5, 7, 9, 10, 48, 50, 51, 53, 365^» Koll. Zeitschr. 4, 169 (1909); 5, 318 (1909) Sykes, a. and Fischer, M. H., 334 Szücs, J., 382 Tachau, p., 231, 232 Taenzer and Unna, P. G., 434 Tait, J., 307^" Takahashi, D. and Rona, P., 235* Tamara, 342* Arch. f. exper. Pathol, u. Pharmek. 59, 1 (1908) Tamman, G., 57* Wiedem. Ann. 34, 299 (1888); Zeitschr. f. physikal. Chemie 10, 255ff. (1892) Tangl, f., 18*, 266 in Oppenheimers Handbuch d. Bio- chemie 3, II, p. 20 Tappeiner, 363 Teague, O. and Buxton, B. H., 84*^, 430*2 1) Journ.ofexper.Med.9,No.3(1907) 2) Zeitschr. f. physikal. Chemie 60, 469-506 (1907); 62, 287-307 (1908) (and Field), 205* Thomas, A. W., 3 Hay ward, G., 228 Thompson, d'Arcy W., 281-''" and Walti, 362 Thouery, 363 Tigerstedt, R. a. a., 293 Translator, 3, 12, 26, 33, 40, 41, 42, 48, 54, 71, 86, 98, 109, 120, 122, 137, 143^", 145, 169, 175, 188, 190, 195^", 199, 210, 210-^", 211^", 220-^", 228, 230, 232, 232-^", 234, 247, 252, 261, 276, 281-^", 287, 298B, 299, 300, 308, 310, 323, 324, 325, 335, 336, 338, 339, 350, 351, 354, 359, 361, 365, 371, 377, 378, 379, 383, 388, 404-405, 409, 435 Traube, J., 108, 100, 163, 211, 236*' ^'', 236-^'^, 385* ■^", 386 Arch. f. Anat. u. Physiol. 87 (1867) Pflügers Arch. 105, 541-558; 559- 572 (1904) and Czapek, F., 240, 386 and Kohler, F., 163, 427-'''' Moritz, 57* Trommsdorf, R. and Hahn, 206 Tröndle, a., 245 True and Kahlenberg, 382 TsuBoi, 298 Tyndall, 119, 125 Uhlenruth, B. D., 257* Uhlirz and Landsteiner, 145* Unna, P. G., 361*^ 364*^ 1) Medizin, Rev. 1, No. 2-4 2) Medizin. Rev. Klinik, No. 42, 43 (1907) 3) Arch. f. mikrosk. Anat. (Waldeyer Festschr.) 78 (1911) and Ehrlich, 424 and Golodetz, L., 163, 434 and Taenzer, 434 Van Slyke, Donald, 210-^", 310, 314 Van't Hoff, J. H., 43 Vaughan and Novy, 210 Vegesack, H. von, 73*, 106, 106* Vernon, 240 Verworn, 388 VoELTz, 346*, 347, 348 Arch. f. d. ges. Physikol. 102, 373- 414 VoGT and Klose, 352 Voigt, J., 366, 373, 374 Voigtländer, 106 Volk, 200, 201 and Eisenberg, 200, 200*, 203 and Landsteiner and Eisenberg, 201 Volkmann, A. W., 218* Ber. d. Kgl. sachs. ges. d. Wissensch. (1874) Vorländer and Häberle, 19^" Vries, H. de and Pfeffer, W., 239 Waentig and Steche, 184 Wagner and Determeyer, 343* Walden, p., 57* 456 AUTHORS' INDEX Waldenberg and Höber, A., 295* Wallace, G. B. and Gushing, A. R., 319* Wallerstein, 180 Walti and Thompson, 362 Warburg and Wiesel, 386 Wasielewski, Th. v., 418 Wasserman, 194, 196^", 206, 207, 208 Webster, R. W., 289 Univ. of Chicago Publ. 10 (Dec, 1902) Weevers, Th., 234 Wegelin, G., 4 Weigert, 434 Weil, R., 210 Weill, E. and Achard, 371 and Robin, A.. 371 Weimarn, p. p. von, 71, 73 Weissmann, 375-''" Weleck, St. and Landsteiner, K., 200 Welter, 250 WiDAL, Fernand, 202^", 339 and Gruber, 202^" WiDMARK, E. M., 222*, 292* Skandinav. Arch. f. Physiol. 23, 421- 430 (1910); 24, 13-22 (1910); 24, 339-344 (1911) Wiedemann, E. and Ludeking, Chas., 134*, 137* Wiedemann's Annalen 25, 433 G., 78 Wiegner, G., 346, 348, 349* Wiesel and Warburg, 386 WiLKE, 119 WiLKE-DöRFÜRT and Zsigmondy and Galecki, 98 Williams, Mattieu, 169^" Winkel, R., 123 WiNKELBEECH, 3S* Winter, 329 WiSLiCENUs, H., 110, 112^"-, 169*, 248*, 249, 250, 258*i, 258^^^- 1) Papier-Ztg. 16 (1910) 2) Tharandter forst. Jahrt. 60, 313- 358 3) KoU. Zeitschr. 6, 17, 87 (1910) and Muth, W., 112* CoUegium No. 255, 256 (1907) Wistinghausen, 63 Witte, 35 Wittka and Pfeiffer, 156 WÖHLER, L., 73* Woodyatt, R. T., 228, 228^" and Fischer, M. H., 335 Wright, 288 Yamanouchi and Lavaditi, 208 Ybrsin and Roux, 110, 198* Young, W. J. and Harden, A., 191* Zangger, H., 55, 56, 63, 109, 172, 173, 173*2, 196^" 1) Ergebnisse d. Physiol. VII (1908) 2) Schweizer. Arch. f. Tierheilkunde 5 (1908) Zeiss, Carl, 120 ZiEGLER, J., 10, 34, 73, 106, 115*2, 138*2, 147^^*, 148*3 and Bechhold, H., 10*, 35, 54*2, 55^ 55*2, 57*1^ 73^ 106, 138*2, 147^", 148*3, 162*2, 238*1, 243, 244, 260*i, 269*3, 319, 320*2, 327, 329, 343, 378*3, 380, 403*2 Kurt, 231 ZiLLESSEN, H. and Araki, F., 226 Zlobicki, 135*, 137* BuU. de I'Acad. de Science de Cracor 488 (1906) ZOTT, 63 Zsigmondy, R., 5*, 6, 10, 41, 48-''", 49, 49^", 73*2, 75^ 75*2^ 77^ 83, 85, 92*3, 93"'-, 98, 99, 122, 125, 143, 342 1) Liebig's Ann. 301, 39 (1898) 2) Z. Erkenntris d. KoU. (Jena, 1905) 3) KoU. Zeitschr. 8, 123 (1911) and Schulz, Fr. N., 143 and Siedentopf, 6, 75 Wilke-Dörfürt and Galecki, 98 Alexander's translation, 247 ZuNTZ, N., 216 and Schumberg, 216 ZuNZ, E. and Jacque, L., 199* Edgar, 86, 89^", 167*', 205, 363 Bull, de la Soc. de Sc. med. et nat. de Bruxelles 67, 178-179 (1909) SUBJECT INDEX Abrin, 204 Absorption, 316 alimentary, 317 of exudates, 323 of fats, 247 influence of protective colloids on, 247 mechanism of, 320, 322 parenteral, 323 percutaneous, 324 of water and crystalloids, 318 Acetic acid, 422 Acids, as fixatives, 421 Acid, poisoning with, 301-310 Adjective dyeing, 430 Adsorption, 19, 21, 22, 84, 109 abnormal, 27 affinitive cm'ves, 25 apparatus, 112 arsenious acid in iron hydroxid gel, 27 change induced by, 109 detergents, 28 determination of electric charge by, 112 and disinfectant action, 398, 399, 400 of dyes, 25 dyeing, 27 effect of on membranes (see Pro- teins), 58 of enzymes, 185 equilibrium, 27-111 gas exchange, 309 influence of, in hemoglobin, 305, 306 grapliic of, 24 in microscopic staining, 425 and immunity reactions, 198 influence of chemical composition, 27 et seq. mechanical, 26 of narcotics, 389 negative, 19-24 negative, simulation of, 27 pharmacological, 363 reversible, 199 Adsorption, saturation, 26 selective, 32 specific, 198 by starches, 135 therapy, 353, et seq. by ultrafilters, 101 and urine secretion, 337 Age, influence of, 55 influence of, on meat, 169 et seq. Agglutinin, 194, 195, 201, 202 Albmnins, 146 acid, 152 alkali, 153 amphoteric, 154 coefficient of diffusion of, 146 electrolyte free, 147, 156 influence of inorganic hydrosols on, 156 Albumin, in milk, 349 as sols, 147 Albuminoids, 161 Albumoses, 166 Alcohol, as fixative, 423 effect on colloids after ingestion, 390 Alcoholism, 324, 325 Aluminimn, 382, 383 Amboceptor, 195-200, 201, 206 Ameba, phagocytosis by, 285 migration of, 284 Anesthetics, 385 Anaphylatoxin, 209 Anaphylaxis, 209 and heavy metals, 382 Anion and cation influence, table, 297 purgative action of, 412 Antagonism of diuretics and narcotics, 411 Antagonism of ions, 82 of salts physiological, 379 Antibodies, 195-197 Anti-enzymes, 191 antigens and immune substance, 205 Antigens, 196 Antitoxins, 191, 195 457 458 SUBJECT INDEX Anthrax, disinfectant action on, 396, 397 inhibition of, growth of, 407, 408 Artefacts, 264 Assimilation, 245 Astringents, 414 action of, 383 Avogadro's law, 43 Bacterial staining, 435 Balneology, 414 Beer, 179 cloudiness of, 180 fermentation of, 181 protective colloids of, 181 Bile, 330 Bio-colloids, 129 Biological determination of adsorption in disinfective, 399 Blood, 299 reaction of, 301 corpuscles, 204, 244 corpuscles, 303 corpuscles, composition of, 304 corpuscles, influence on viscosity, 314-315 corpuscles, hemolj^sis (see also Wasserman reaction), 244 corpuscles, osmotic pressure of, 304 corpuscles, structure of, 305 Blotting paper, disinfectant testing with, 408 Boyle's law, 51 Bread, 177 action of hemaglobin in, 308 et seq. gluten restoration of, 177 staleness of, 178 war, 178 Bromin, therapeutic action of, 381 Bronchial glands, 328 Brownian Zsigmondy movement, 49- 53 Bubble method, of examining milk, 173 Buffer substances of blood, 300 Butter, 175-346 Calcium, action of, 298A, 379-381 compounds, colloidal preparations of, 381 condition of, in serum, 302 ion, influence on phagocytosis, 287 Calcium, phosphate, condition of, in milk, 349 utilization, 325, 361 Calculi, urinary, 343 Carbohydrates, 133 Carrel-Dakin disinfection, 405 Casein, 163, 348 Cations as diuretics, 409 Casts, urinary, 344 Carbon dioxid, influence of, on m-ine excretion, 333 solubility in blood, 309, 310 Catalysers, 31, 183 Catalysis, 81 Cell, 276 membrane, 279 et seq. structure of, 276 cerebrospinal fluid, 354 Cheese, 176 Chemical attraction of precipitates, 260 Chemical combination, 19, 22 determination of adsorption in dis- infection, 398 theory of dyeing, 425, 426 Chemotaxis, 286 Chloroform poisoning, delayed, 389 distribution of, in disinfection, 398 Cholesterin, 87, 140, 141 Chromic acid, 421 Circulation, of crystalloids, 235, 238 coUoids, 239 of gases, 235 of material, 235 of water (see also swelling), 236 et seq. Clotting of blood, 300 Cloudy swelling, 228 Coagulation, 82, 114, 142, 149 of blood, effect of gelatin on, 365 chemical, 143, 149 fractional, 82 by freezing, 144 by heat, 142, 151 irreversible, 143 by light, 144 Co-enzymes, 191 Collagen, 161 Colloidal protection, in milk, 349 in urine, 343 Colloids, aging of, 72, 74 artificial, 5 SUBJECT INDEX 459 Colloids, crystallization of, 71 color of, 5, 7 consistency of, 64 death of, 73 definition of, 3 dynamic balance in organism, 308 electrolytes, 46 electrical properties of, 77 electrical production of, 4 hydrate, 36 hydrophile, 9 hydrophobe, 9 importance in body, 126 et seq. intravenous action of, 365 life curve of, 72 et seq. mechanical production of, 4 migration of, 84, 86 optical properties of, 75 particle size, 7 pharmaceutical action of, 362 protective, 7, 11, 36, 77, 86, 181, 203, 363 protective, in urine, 343 Colloid swelling and urine secretion, 323, 336 swelling, and equilibrium, 341 ■Colloidal properties of dye mixtures, 430 antimony, 377 arsenic, 377 mercury, 376 phosphorus, 377 metals, therapeutic use of, 365 silver, 366 silver, distribution of, 373 silver, effect on blood, 371 suver, effect on temperature, 373 silver, therapeutics of, 374 silver, in infections, 376 silver, in pneumonia, 375 silver, in wounds, 366 silver, protective colloids for, 366 sulphm-, 376 swelling state of tissues, 415 Complement deviation or fixation (see also Wassermann reaction), 194, 206, 207, 208 Complement, 189, 198, 200, 206 Concentration, by absorption, 27 couples, 62 ■Conduction by nerves, 354 Cooperation of drugs, 361 Cream, 175, 346 artificial, 175 sophistication of, 175 Critical narcotic concentration, 385 Cryoscopy, 341 Crystals, force producing, 17 Dehydrating action of purgatives, 414 Dehydration of food, 169 Development, 252 Diabetes insipidus, 341 Dialysis, collodin sac, 91 methods, 89 et seq. Diaphragms, charge of, 78 Diarrhea, 321, 323 treatment of, 364 Difl^usion, 103 et seq. apparatus, 105, 106 coefficient of, 45, 52, 190 influence of adsorption upon, 55 influence of substances on, 55 in jellies, 54 relation to dyeing, 428 of protein, 146 DigestibUity of mük, 349, 350 Digitalis, 411 Disinfectants and dissociation, 402 Disinfectant action of cresol, 396 action of H and OH ions, 394 action of phenyl group and halogens, 394 action of sulpho groups, 394 and adsorptive capacity, 394, 395, 396, 397 action of chloroform, 398 action and dilution, 396 and death, 396 and inhibition, 396 of specific character, 394 salts of heavy metals, 394 Disinfectants, testing oi, 405, 406 Disinfection, 359 definition of, 391 of the skin, 395 mechanism of, 391 and permeability, 402 and surface tension, 391 Dispersed phase, (see Phase) 5, 11, 12 Dispersion, 3 Dissimilation, 245 460 SUBJECT INDEX Dissimilation, influence of enzymes, 250 Dissociation and disinfection, 402 Distribution, 20, 22, 31 et seq. of disinfectant, 398, 399 of disinfectant on microorganism, 393 Henry's law, 20 in toxicology, 360 Diuretics, 409 Diuretic action of chloral caffeine, 362 of caffeine, 411 of salts administered intravenously, 410 of theobromin, 411 of urea, 411 Double staining, 434 Drugs, influence of, on kidneys, 338 Dyeing (see Staining), 200-206, 407 Eczema, 234 Edema, 223 et seq., 377 controversy concerning, 229 et seq. of the brain, 231, 352 treatment of, 339 Elastic fibers, 434 Electric charge, enzymes, 187 migration, 118 migration, apparatus for study of, 118 Electrodes — nonpolarizable, 119 Electro-endosmosis, 78 Electrolyte (see also Salt) Electrolytes, 149 influence of, on viscosity of gelatin, 162 Emulsion, definition of, 5 Emulsions, formation of, 140 Enzymes, 182 adsorption analysis of, 185 aging of, 188 colloidal nature of, 183 diffusion coefficient of, 190 electric charge of, 187 inactivation of, 189 purification, 187 specification of, 188 synthesis by, 188 ultrafiltration of, 190 Equilibrium, 341 emulsion, 38 irreversible, 28 Equilibrium, reversible, 28 study of by ultrafiltration, 102 Erythrocytes (see Blood corpuscles, volume of), 307 Excretion, 326 Exudate, 223 Fatty degeneration, nature of, 377 Fats, deposition of, 246 resorption of, 246 Fibrin, 160 Flocculation, 83, 86, 117 vs. salting out, 83, 87 Flour, 176 Foods, 168 Ferric hydroxid, negative coUoid, 384 positive colloid, 384 Ferric oxid as arsenic antidote, 385 intravenous injection of, 384 negative, 384 positive, 384 Fixatives, 420 Fixing and hardening of tissues, 419^ 420 Formaldehyde, distribution of, in dis- infection, 398 as fixative, 423 Freezing, 216 Freezing-point depression, in milk, 340 in urine, 351 Friction internal (see Viscosity), 64 Gas exchange (see Respiration) Gastric juice, 329 Gay-Lussac's Law, 51 Gel, definition of, 4, 8 elastic, 66 freezing and thawing of, 66 Gelatin, 161 Gelatinization, 161 et seq. time of for agar, 138 Gibbs' Theorem, 25 Glaciation, 216 Gland, 326 Glaucoma, 227 Globulins, 158 artificial, 159 Gold figure, 85 Glycogen, staining of, 433 Golgi's stain, mechanism of, 431 Gout, 148 SUBJECT INDEX 461 Gram's stain, 30, 435 Growth, 252 biological, with shrinking, 265 Growth of plants, 2()7 influence of chemical reaction, 267 Hardening, histological, 424 Heavy metals and salts of, 157 Heavy metals, as £_>catives, 421 Heavy metal specificity, 383 Hematin, 165 Hemoglobin, 164 function of, 309 (see also Blood corpuscles, respiration) synthetic, 385 Hemolysis, 305, 306 induced by colloids, 371 Hemolysins, 196, 200 Honey, 176 H ion concentration in blood, 302 influence upon circulation, 314 influence upon erythrocytes, 312 influence of CO2 upon, 311 Histone, 160 Hydremia, 341 Hydrosol, 11 Immune substances, 201 Immunity reactions (see also Precipi- tin), 193 et seq. Inactivation, by shaking, 34 Inflammation, 232 Inhibition, 354 zones or irregular series, 84, 118, 140 Instant values, 51 Interferometer, 120 Interface (see Surface) Interface, 14 Internal friction, 113 of albumin, 152 of globulin, 159 of gums, 137 Intestinal inflammation, 322. Intestinal secretion, 330 Inulin, 133, 210 lodin, therapeutic action of, 380 Iron, astringent action, 383 as arsenic antidote, 384 action ot colloidal, 383 colloidal preparations, value of, 383, 384 Iron, mechanism in hemostasis, 384 intravenous injection of, 384 oral administration of, 384 pharmacological action of, 383 Irregular series, 157, 203 Irritability of nerves, 353 Isoelectric zone, 77, 84, 161 point for casein, 164 hemoglobin, 165 Jellies, structure of, 9 Keratins, 163 Kinetic theory, 50 Lanolin, composition of, 416 properties of, 416 Layered structm-es, 261 et seq. Lecithin, 87, 139, 140 precipitation of magnesium salts, 388 Liesegang' s rings, 263 Lipoids, definition of, 139 reaction with narcotics, 389 staining of, 433 Local anesthetics, mechanism of, 389 Lymph, 303 Lyotropic series, 140 Lyotropism, definition, 81 Maceration for microscopic study, 419 Magnesium sulphate, pm-gative action of, 413 Margarine, 175 Mass staining, 431 Meat, 169 boiling, 172 cold storage, 169 preserving, 172 Meiostagmin reaction, 211 Melting temperature, 114, 138, 161 Membrane, definition of, 56 formation of, 56 growth, 57 self regulation, 58 semipermeable, 57, 236 and crystalloids, 58 equilibria, 59 hydrolysis, 59 and substances interchange, 239 Membranes, mfluence on substance in- terchange, 239 462 SUBJECT INDEX Membranes and electrolytes, 59 Mercuric chlorid poisoning, 143 Mercury poisoning, 383 salts, as disinfectant, 400 Metamorphosis, 252 Microorganisms, influence of suspen- sions on growth, 393 Microscopic technic, 417 Milk, adulteration of, 174 colloids of, 174, 349 condensed, 174 cows and woman's, 349, 351 dried, 351 food accessories of McCollum, 351 freezing point depression of, 351 homogenized, 346, 348 raw and boiled, 350 surface pellicles in, 346 skin formation in, 350 ultrafiltration of, 350 ultramicroscopy of, 349 viscosity of, 173, 348 drying of, 174 examination of, 173 ultrafiltration of, 174 Molecule, 42 Molecular, movement, 50 weight, 42 et seq. Mordants, 431, 434 Movement of organisms, 282 Myosin, 160 Mucins and mucoids, 165 Muscle, 289 et seq. influence of Ca on, 298a Muscles, influence of drugs on, 298a influence of lactic acid on, 298 electric phenomena of, 294 fatigue, 291 function, 292, 296, 298 influence of electrolytes on, 294 influence of phosphates on, 298a working model of, 296 Narcosis, distribution of narcotic, 388 and oxygen absorption, 388 and respiration, 388 and change in turgor, 387 Narcotics, 385 action and electric conductivity, 386 - action and inhibition of hemolysis, 386 Narcotics, action of, on permeation of electrolytes, 389 effect of, on plasma pellicle, 389 lipoid solution theory, 385 Meyer-Overton theory, 385 Narcotics, toxic action of, 390 and disturbance of oxidation, 390 action on catalase, 390 Narcotic action and inhibition of fer- ment action, 386 and inhibition of oxidation, 386 action and lipoid solubility, 386 action of magnesium salts, 386, 388 action and protein precipitation, 386 action and plasma pellicle, 386 action and surface tension, 386 Nephelometer, 120 Nephritis, 338 Nerves, 352 Neurobiotaxis, 261 Neutral salts toxic action, 379 Nucleins, 160 Nucleoalbumins, 163 Nucleus, 267, 279 staining of, 433 Organism, circulation of, 23 as colloid, 213 Optical methods, 119 rotation, 156 Osmic acid, 421 Osmotic compensation method, appifeCi to milk, 348 growths, 250 pressure, measurement of, 43, 51, 10 1 instruments, 106 compensation method, 107 measurements of starches, 135 Ossification, 268 theories of, 269 Oxygen and air as disinfectants, 399 Pancreatic juice, 330 Parthenogenesis, 245 Particle size (see also Proteins), 41 Pellicle, plasma, nature of, 240 et seq. of blood corpuscles, 301 Peptisation, 84 Permeability, of cell membrane, 242, 243 chemical regulation of, 244 SUBJECT INDEX 463 Permeability, lethal change of, 245 photo regulation of, 245 selective, 63 Peptones, 166 Phagocytosis, 285, 287 et seq. induced by colloidal silver, 372 influence of cation, 287 Phase, definition of, 5 influence of serum, 288 Phenol, distribution of, in disinfection, 398 Photodromy, 76 Picric acid, 422 after vital staining, 422 Plasma, 299 Plasma pellicle, effect of narcotics on, 389 Plasmolysis, of aspergillus cells (see also Hemolysis), 241, 243 Poisons, 360 Powders, therapeutic effect of, 364 Precipitation, 80 et seq., 157 (see also Coagulation) by alcohol, 152 influence of electrolytes, 163 irreversible, 87 of gelatin, 162 zones, 157 Precipitin reaction, 56 (see also Immunity reaction, 201, 202) Pro-enzymes, 191 Protamine, 160 Protection, preferential, 371 Protective ferment, 210 Proteins, 142 adsorption phenomena, 145 et seq. crystallization of, 144 ultrafiltration of, 146 ultramicroscopic studies of, 144 Proteoses, 167 Protoplasm, 277 et seq. death of, 279 staining of, 433 Pulsation, influence of, on secretion, 327, 332 Purgatives, 409 mechanism of, 412 Pyrosol, 11 Radioactivity, therapeutic action, 415 Radioactive substances, 87 Regulation, automatic of cell metab- olism, 244 Respiration (see Gas exchange) Rhythmic phenomena, 263 Ricin, 204 Rigor Mortis, 291 Saccharo-coUoids, 133 Saliva, 328 Salts, concentration of in intestines, 412 Salt, distribution of, 233 Salting out, 80 Saponin, 34 Salts as fixative, 422 therapeutic and toxic action of, 378 Salves, mechanism of, 416 Secretion, 315 ei seq. Section staining, 431 Separation, by shaking out the foam, 35 Serum, content of globulin, 159 influence of on solubility of salts, 302 surface tension of, 303 Shrinking, 65 Silicic acid jelly, size of pores, 10 Silk threads, disinfectant testing with, 406 Silver nitrate, disinfectant action of, 39S Size, therapeutic, use of, 365 Skin disinfection, 404 Soap as a disinfectant, 404 Sol, definition of, 3 Solidification temperature, 114, 161 Solubuity, influence of gelatin on, 161 selective, 63 Solutions, homogeneous, 6, 19 Specific action of disinfectants, 401, 402 Specific chemical action of disinfectant, 401 Spongin, 163 Staining, 424 bacterial, 435 double, 434 method of, 431 et seq. technic of, 431 theory of, 424 et seq. Starch, 133 agar, 136 cellulose, 138 crystaUizable, 135 glucosides, 136 464 SUBJECT INDEX Starch, granules, relation to mineral content, 134 gums, 136 soluble, 134 paste, 134, 135 molecule, size of, 135 Structures, genesis of, 259 osmotic, 253 et seq. Surface, development of, 13 phenomena of, 11 pellicles, in milk, 346 tension, 287 tension, and disinfection, 391 tension, measiu-ement of, 108 tension, and muscular action, 296 tension, by ultrafiltration, 16 Surface tension, 14, 81 in common things, 34 skins, 33 of solids, 18 in stains, 34 in milk, 173 Suspension, definition of, 5 Sweat glands, 345 Swelling, 65, 169, 173 influence ci electrolytes on, 67 et seq., 222 and intestinal absorption, 320 of gelatin, 161 of organs, 217 measurements of, 114 • pressure, 66, 115 range, 217, 219 ratio, 217 Threshold "electrolyte," 83, 118 Tissue behavior with dyes and fixa- tives, 433 growth in vitro, 246 Tropisms, 283 Toxin-antitoxin, 31 Tropisms, explanation of, 283 Turgor and irritability, 387 Tyndall phenomenon, 75 Ultracentrifugation, 6, 42 Ultrafilter, theory of dyeing, 427 theory of plasma pellicle, 241 Ultrafiltration, 6, 10, 42, 58 absorption in, 101 applications of, 102 Ultrafiltration, gauging of, 99 methods of, and apparatus, 95 et seq. of albumoses, 166 of cerebrospinal fluid, 354 of enzymes, 190 of milk, 174 of proteins, 144 of viruses, 393 Ultramicroscope, 6, 75, 122 et seq. cardioid condenser, 126 Ultramicroscopy of milk, 349 Urea, diuretic action of, 334 • influence on diffusion, 55 Ultrafiltration, 327, 332 of urine, 332 Urine, colloids of, 342 concentration of, 335, 336 effect of drugs on, 338 freezing point depression, 340 glomerular secretion of, 331 et seq. normal, 342 pathological excretion of, 338, 343 secretion of, 323, 330, 332, 336 surface tension of, 343, 344 threshold, substances in, 336 Viscosity, 113 of albumins, 151, 154 of blood, 310, 311, 312 of blood and diuresis, 334 Viscosity of plasma, 311 Viscosimeters, 113 Viscosity, influence of caffeine on, 152 influence of neutral salts on, 152 Viscosity of serum, 302 Vital staining, 431 Wassermann reaction (see Comple- ment deviation) Water, in blood, 220 content influence on muscle function, 298 distribution in body, 415 distribution of, in organism, 215 in muscles, 220 in organs (see Swelling), 217, 218, 219, 221 pathology of, 223 of solution, 66 of swelling, 66 Wood formation, 248 et seq. Wound disinfection, 404, 405 D. VAN NOSTRAND COMPANY 25 PARK PLACE NEW YORK SHORT-TITLE CATALOG OF |3ublication0 anö Jmportation© OF SCIENTIFIC AND ENGINEERING BOOKS This list includes the techaical publications of the following English publishers: SCOTT, GREENWOOD & CO. JAMES MUNRO & CO., Ltd. CONSTABLE &COMPANY.Ltd. TECHNICAL PUBLISHING CO. BENN BROTHERS, Ltd. for whom D Van Nostrand Company are American agents. December, 1919 SHORT=TITLE CATALOG OF THB Publications and Importations OF D. VAN NOSTRAND CO/V\PANY 25 PARK PLACE, N. Y. Ail Prices in this list are J^BT. All bindings are in cloth unless othertvise noted. Abbott, Ai V. The Electrical Transmission of Energy. .8vo, *$5 oo A Treatise on Fuel , . i§mo, o 75 Testing Machines i6mo, o 75 Abraham, Herbert. Asphalts and Allied Substances 8vo, 5 00 Adam, P. Practical Bookbinding i2rao, 2 50 Adams, H. Theory and Practice in Designing 8vo, *2 50 Adams, H. C. Sewage of Sea Coast Towns 8vo, *2 50 Adams, J. W. Sewers and Drains for Populous Districts 8vo, 2 50 Addyman, F. T. Practical X-Ray Work 8vo, 5 oo Adler, A. A. Theory of Engineering Drawing 8vo, 2 50 Principles of Parallel Projecting -line Drawing ...8vo, i 25 Aikman, C. M. Manures and the Principles of Manuring 3vo, yKefrmtmg.) Aitken, W. Manual of the Telephone Svo, *8 00 d'Albe, E. E. F. Contemporary Chemistry i2mo, i 50 Alexander, J. Colloid Chemistry lamo, i 00 Allan, W. Strength of Beams Under Transverse Loads i5mo, 075 Theory of Arches i6mo, Allen, H. Modern Power Gas Producer Practice and Applications . : 2010, ^Reprinting.) Anderson, J. W. Prospector's Handbook . izmo, i 75 Andes, L. Vegetable Fats and Oils ävo, *6 00 Animal Fats and Oils 8vo, 5 00 Drying Oils, Boiled Oil, and Solid and Liquid Driers Svo, *6 00 Iron Corrosion, Anti-fouling and Anti-corrosive Paints 8vo, 6 00 Oil Colors, and Printers' Ink 3vo, 4 00 Treatment of Paper for Special Purposes izmo, 3 00 Andrews, E. S. Reinforced Concrete Construction i2mo, *2 00 Theory and Design of Structures 8vo, *3 50 Further Problems in the Theory and Design of Structures. .. .8vo, *2 50 — The Strength of Materials 8vo, *4 00 Elastic Stresses in Structures 8vo, 9 00 Andrews, E. S^ and Heywood, H. B. The Calculus for Engineers. lamo, *2 00 Annual Reports on the Progess of Chemistry. Fifteen Volumes now ready. Vol. I., 1904, Vol. XV., 1919 Sva^ each, 2 00 Argand, M. Imaginary Quantities i5mo, o 75 Armstrong, R., and Idell, F. E. Chimneys for Furnaces and Steam Boilers. lömo, o 75 2 50 2 oo 2 50 I 25 I 50 0 75 0 75 I 50 2 00 D. VAN NOSTRAiND CO.'S SHORT TITLE CATALOG 3 Arnold, E. Armature Windings of Direct-Current Dynamos 8vo, 2 00 Asch, W., and Asch, D. The Silicates in Chemistry and Commerce. 8vo, 7 50 Ashe, S. W., and Keiley, J. D. Electric Railways. Theoretically and Practically Treated. Vol. I. Rolling Stock i2mo, *2 50 Ashe, S. W. Electric Railways. Vol. II. Engineering Preliminaries and Direct Current Sub-Stations 12 mo, Electricity: Experimentally and Practically Applied i2mo, Ashley, R. H. Chemical Calculations i2mo, Atkins, W. Common Battery Telephony Simplified i2mo, Atkinson, A. A. Electrical and Magnetic Calculations 8vo, Atkinson, J. J. Friction of Air in Mines i6mo, Atkinson, J. J., and Williams, Jr., E. H. Gases Met with in Coal Mines. i6mo, Atkinson, P. The Elements of Electric Lighting lamo, The Elements of Dynamic Electricity and Magnetism 12 mo, Auchincloss, W. S. Link and Valve Motions Simplified 8vo, *i 50 Audley, J. A. Silica and the Silicates 8vo (In Press.) Austin, E. Single Phase Electric Railways 4to, *5 00 Austin and Cohn. Pocketbook of Radiotelegraphy (In Press.) Ayrton, H. The Electric Arc 8vo, 5 50 Baff, W. E. Sale of Inventions i2mo (In Press.) Bailey, R. D. The Brewers' Analyst Svo, *5 00 Baker, A. L. Quaternions 8vo, i 50 Thick-Lens Optics izmo, *i 50 Baker, Benj. Pressure of Earthwork i6mo, Baker, G. S. Ship Form, Resistance and Screw Propulsion 8vo, *4 50 Baker, I. 0. Levelling i6mo, o 75 Baker, M. N. Potable Water i6mo, o 75 Sewerage and Sewage Purification i6mo, o 75 Baker, T. T. Telegraphic Transmission of Photographs izmo, (Reprintinih^ Bale, G. R. Modern Iron Foundry Practice. i2mo. Vol. I. Foundry Equipment, Materials Used *3 00 Ball, J. W. Concrete Structures in Railways 8vo, *2 50 Ball, R. S. Popular Guide to the Heavens 8vo, *5 00 ■ Natural Sources of Power 8vo, 2 50 Ball, W. V. Law Affecting Engineers 8vo. *3 50 Bankson, Lloyd. Slide Valve Diagrams ._. i6mo, o 75 Barham, G. B. Development of the Incandescent Electric Lamp.Svo, 2 50 Barker, A. F. Textiles and Their Manufacture 8vo, 2 50 Barker, A. F., and Midgley, E. Analysis cf Woven Fabrics 8vo, 3 50 Barker, A. H. Graphic Methods of Engine Design i2:no, 2 00 Heating and Ventilation 4to, 9 00 Barnard, J. H. The Naval Militiaman's Guide i6mo, leather i 00 Barnard, Major J. G. Rotary Motion i6mo, o 75 Barnes, J. B. Elements of Military Sketching i6mo. *o 75 Barnett, E. deB. Coal-Tar Dyes and Intermediates 8vo, 3 50 Explosives, Matches and Pyrotechny 8vo (/;; Press.) ■ Synthetic Dyes 8vo (In Press.) Barrowcliff, M., and Carr, F. H. Organic Medicinal Chemicals. .8vo, (lit Press.) 4 D. VAN NOSTRAND CO.'S SHORT TITLE CATALOG Barrus, G. H. Engine Tests 8vo, Baterden, J. R. Timber b\o, Bates, E. L., and Charlesworth, F. Practical Mathematics and Geometry i2mo, Part I. Preliminary Course Part II. Elementary Course Part III. Advanced Course Practical Mathematics i2mo, Practical Geometry and Graphics lamo, Batey, J. The Science of Works Management i2mo, Steam Boilers and Combustion i2mo, Bayonet Training Manual i5mo, Beadle, C. Chapters on Papermaking. Five Volumes i2mo, each, Beaumont, R. Color in Woven Design 8vo, Finishing of Textile Fabrics 8vo, Standard Cloths 8vo, Beaumont, W. W. The Steam-Engine Indicator 8vo, Bechhold, H. Colloids in Biology and Medicine 8vo, Beckwith, A. Pottery 8vo, paper. Bedell, F. Airplane Characteristics 8vo, ^'The Air Propeller 8vo, The Airplane 8vo (/;; Press.) Bedell, F., and Pierce, C. A. Direct and Alternating Current Manual. 8vo, Beech, F. Dyeing of Cotton Fabrics , 8vo, Dyeing of Woolen Fabrics 8vo, Beggs, G. E. Stresses in Railway Girders and Bridges (In Press.) Begtrup, J. The Slide Valve 8vo, Bender, C. E. Continuous Bridges ' H5mo, — —Proportions of Pins Used in Bridges i6mo, Bengougfa, G. D. Brass {In Press.) Bennett, H. G. The Manufacture of Leather 8vo, 6 oo .^-^. — Animal Proteids Svo, (In Press.) Ber»thsen, A. A Text-book of Organic Chemistry i2mo, 3 50 Berseb, J. Manufacture of Mineral and Lake Pigments 8vo, 6 00 Beveridge, J. Papermaker's Pocket Book i2mo, *4 ou Binnie, Sir A. Rainfall Reservoirs and Water Supply 8vo, 4 00 Binns, C- F. Manual of Practical Potting 8vo, 6 00 . The Potter's Craft i2mo> *2 00 jBirchmcffe, W. H. Interpretation of Gas Analysis i2ino, *i 25 Blaine, R. G. The Calculus and Its Applications i2mo, *i 75 Blake, W. H. Brewers* Vade Mecum 8vo, *4 00 Blanchard, W. M. Laboratory Exercises in General Chemistry. .i2mo, i 00 Blasdale, W. C. Quantitative Chemical Analysis izmo, 2 50 Bloch, L. Science of Illumination 8vo, 2 50 Blyth, A. W. Foods: Their Composition and Analysis Svo, 8 50 — — Poisons: Their Effects and Detection 8vo, 8 50 Böckmann, F. Celluloid i2mo, *2 50 B(?4iner, G. R. Hydraulic Motors and Turbines i2mo, 5 00 Koileau, J. T. Traverse Tables 8vo, 5 00 Bonney, G. E. The Electro-platers' Handbook i2mo, i 50 ■4 00 *2 50 I 00 I 00 I 50 *2 00 2 00 *2 00 *2 GO 0 30 *2 00 *6 00 *5 00 *6 00 2 50 5 00 0 60 I 60 I 00 2 00 5 00 *3 50 ^2 00 0 75 0 75 2 50 *I 00 I 00 0 75 =■6 00 o 75 6 00 5 00 0 75 I 75 2 25 3 00 2 50 *2 25 *4 25 2 00 D. VAN NOSTRAND CO.'S SHORT TITLE CATALOG 5 Booth, N. Guide to the Ring-spinning Frame i2mo, *?. 00 Booth, W. H. Water Softening and Treatment. .. .8vo (Reprinting.) Superheaters and Superheating and Their Control 8vo, 'i 50 Böttcher, A. Cranes: Their Construction, Mechanical Equipment and Working 4to ( Reprinling. )■ Bottler, M. Modern Bleaching Agents. . .- i2mo, Bottone, S. R. Magnetos for Automobilists i2mo, Electro-Motors, How Made and How Use i2mo, Boulton, S. B. Preservation of Timber i6rao, Bourcart, E. Insecticides, Fungicides and Weedkillers 8vo, Bourgougnon, A. Physical Problems i6mo, Bourry, E. Treatise on Ceramic Industries 8vo, Bowie, A. J., Jr. A Practical Treatise on Hydraulic Mining 8vo, Bowls, 0. Tables of Common Rocks i5mo, Bowser, E. A. Elementary Treatise on Analytic Geometry i2mo, Elementary Treatise on the Differential and Integral Calculus . 1 2mo, Elementary Treatise on Analytic Mechanics i2mo, Elementary Treatise on Hydro-mechanics i2mo, — ■ — A Treatise on Roofs and Bridges i2mo, Boycott, G. W. M. Compressed Air Work and Diving Svo, Bradford, G. Whys and Wherefores of Navigation i2mo, Sea Terms and Phrases i2mo, fabrikoid {In Press. ) Bragg, E. M. Design of Marine Engines and Auxiliaries Svo, 4 00 Brainard, F. R. The Sextant i6mo, Brassey's Naval Annual for 1919 Svo, 10 00 Briggs, R., and Wolff, A. R. Steam-Heating i6mo, o 75 Bright, C. The Life Story of Sir Charles Tilsoa Bright Svo, *4 50 Telegraphy, Aeronautics and War 8vo, 6 00 Brislee, T. J. Introduction to the Study of Fuel. .8vo (Reprinting.) Broadfoot, S. K. Motors: Secondary Batteries i2m.o, o 75 Broughton, H. H. Electric Cranes and Hoists Brown, G. Healthy Foundations i6mo, o 75 Brown, H. Irrigation .8vo (Reprinting.) Brown, H. Rubber 8vo, W. A. Portland Cement Industry 8vo, Brown, Wm. N. Dipping, Burnishing, Lacquering and Bronzing Brass Ware i2mo, Handbook on Japanning i2mo. Brown, Wm. N. The Art of Enamelling on Metal i2mo, ■ ■ House Decorating and Painting i2rao, History of Decorative Art i2mo Workshop Wrinkles Svo, Browne, C. L. Fitting and Erecting of Engines Svo, Browne, R. E. Water Meters i6mo, Bruce, E. M. Detection of Common Food Adulterants i2mo, Brunner, R. Manufacture of Lubricants, Shoe Polishes and Leather Dressings 8vo, Buel, R. H. Safety Valves iGmo, Bunkley, J. W. Military and Naval Recognition Book i6mo, Burley, G. W. Lathes. Their Construction and Operation i2mo, Machine and Fitting Shop Practice. 2 vols i2mo, each, Testing of Machine Tools i2mo, Burnside, W. Bridge Foundations ' i2mo, 2' 5'> 3 00 *i 50 "•a 00 *2 00 ""'2 00 *o 50 --I 00 *I 50 0 75 I 40 3 50 0 75 I 00 2 00 2 00 2 00 *2 00 (j D. VAN NOSTRAXD CO.'S SHORT TITLE CATALOG Lurstall, F. W. Energy Diagram for Gas. With Text 8vo, 150 Diagram. Sold separately *i 00 Burt, W. A. Key to the Solar Compass i6mo, leather, 2 50 Buskett, E. W. Fire Assaying lamo, *i 25 Butler, H. J Motor Bodies and Chassis 8vo, *3 00 Byers, H. G., and Knight, H. G. Notes on Qualitative Analysis. . . .8vo, (At'ü,' Edition i)i Preparation.) Caiö, W. Brief Course i» the Calculus i2mo, *i 75 Elastic Arches i6mo, o 75 ■- Maximum Stresses i6mo, o 75 Practical Designing Retaining of Walls i6mo, 075 , Theory of Steel-concrete Arches and of Vaulted Structures. i6mo, o 75 Theory of Voussoir Arches i6mo, o 75 Symbolic Algebra i6mo, o 75 Calvert, G. T. The Manufacture of Sxilphate of Ammonia and Crude Ammonia lamo, 4 00 Camm, S.^ Aeroplane Construction i2mo, 3 00 Carhart, H. S. Thermo Electromotive Force in Electric Cells, (Jn Press.) Carey, A. E., and Oliver, F. W. Tidal Lands. 8vo, 5 00 Carpenter, F. D. Geographical Surveying i6mo, Carpenter, R. C, and Diederichs, H. Internal Combustion Engines. Svo, 5 50 Carter, H. A. Ramie (Rhea), China Grass lamo, =^3 00 Carter, H. R. Modern Flax, Hemp, and Jute Spinning.. Svo, ^-3 50 Bleaching, Dyeing and Finishing cf Fabrics 8vo, *i 25 Gary, E. R. Solution of Railroad Problems with the Slide Rule. . i6mo, *i 00 Casler, M. D. Simplified Reinforced Concrete Mathematics i2mo, ^^i 00 Cathcart, W. L. Machine Design. Part I. Fastenings 8vo, *3 00 Cathcart, W. L., and Chaffee, J. I. Elements of Graphic Statics . . .8vo, *3 00 ■ Short Course in Graphics lamo, i 50 Caven, R. M., and Lander, G. D. Systematic Inorganic Chemistry. i2mo, 2 25 Chalkley, A. P. Diesel Engines 8vo, *4 00 Chalmers. T. W. The Production and Treatment of Vegetable Oils, 4to, 7 50 Chambers' Mathematical Tables 8vo, 2 50 Chambers, G. F. Astronomy i6mo, *i 50 Chappel, E. Five Figure Mathematical Tables Svo, 250 Charnock, Mechanical Technology 8vo, 3 50 Charpentier, P. Timber 8vo, *6 00 Chatley, H. Principles and Designs of Aeroplanes i6mo, o 75 How to Use Water Power i2mo, *i 50 Gyrostatic Balancing 8vo, *i 25 Child, C. D. Electric Arc 8vo, *2 00 Christian, M. Disinfection and Disinfectants i2mo, 2 50 Chtistie, W. W. Boiler-waters, Scale, Corrosion, Foaming Svo, *3 00 ■ Chimney Design and Theory 8vo, *3 00 Furnace Draft i6mo, o 75 . Water: Its Purification and Use in the Industries 8vo, *2 00 Church's Laboratory Guide 8vo, 2 50 Clapham, J. H. Woolen and Worsted Industries 8vo, 200 D. VAN XOSTRAXD CO.'S SHORT TITLE CATALOG 7 Clapperton, G. Practical Papermaking 8vo (Reprinting.) Clark, A. G. Motor Car Engineering. Vol. I. Construction *4 00 Vol. II. Design 8vo, *3 50 Clark, C. H. Marine Gas Engines. New Edition 2 00 Clarke, J. W., and Scott, W. Plumbing Practice. Vol. I. Lead Working and Plumbers' Materials 8vo, *4 00 Vol. II. Sanitar\' Plumbing and Fittings (In Press.) Vol. in. Practical Lead Working on Roofs (In Press.) Clarkson, R. P. Elementary Electrical Engineering (In Press.) Clerk, D., and Idell, F. E. Theory of the Gas Engine i6mo, o 75 Clevenger, S. R. Treatise on the Method of Government Surveying. i6mo, morocco, Clouth, F. Rubber, Gutta-Percha, and Balata 8vo, Cochran, J. Concrete and Reinforced Concrete Specifications 8vo, ■ Treatise on Cement Specifications 8vo, Cocking, W. C. Calculations for Steel-Frame Structures i2mo. Coffin, J. H. C. Navigation and Nautical Astronomy lamo, Colbum, Z., and Thurston, R. H. Steam Boiler Explosions. .. .i6mo, Cole, R. S. Treatise on Photographic Optics lamo, Coles-Finch, W. "Water, Its Origin and Use 8vo, Collins, C. D. Drafting Room Methods, Standards and Forms 8vo, Collins, S. Hoare. Plant Products and Chemical Fertilizers 8vo, Collis, A. G. High and Low Tension Switch-Gear Design 8vo, Switchgear i2mo, Colver, E. D. S. High Explosives 8vo, Comstock, D. F., and Troland, L. T. The Nature of Electricity and Matter 8vo, Coombs, H. A. Gear Teeth i6mo, Cooper, W. R. Primary Batteries 8vo, Copperthwaite, W. C. Tunnel Shields 4to, Corfield, W. H. Dwelling Houses i6mo. Water and Water-Supply i6mo, Cornwall, H. B. Manual of Blow-pipe Analysis 8vo. Cowee, G. A. Practical Safety Methods and Devices 8vo, Cowell, W. B. Pure Air, Ozone, and Water r2mo, Craig, J. W., and Woodward, W. P. Questions and Answers About Electrical Apparatus i2mo, leather, Craig, T. Motion of a Solid in a Fuel i6mo, . Wave and Vortex Motion i6mo, Crehore, A. C. Mystery of Matter and Energy 8vo, . Kew Theory of the Atom (/;.- Press.) Crocker, F. B., and Arendt, M. Electric Motors 8vo, *2 50 Crocker, F. B., and Wheeler, S. S. The Management of Electrical Ma- chinery i2mo, *i 00 Crosby, E. U., Fiske, H. A., and Forster, H. W. Handbook of Fire Protection i2mo, 4 00 Cross, C. F., Bevan, E. J., and Sindall, R. W. Wood Pulp and Its Uses 8vo (R'\hriii!ii!a.) Crosskey, L. R. Elementarv Perspective 8vo, i 50 2 50 '■'6 00 *2 50 *I 00 *2 50 3 GO 0 75 2 00 *5 00 2 GO 3 OG ■■3 50 0 50 12 50 2 50 0 75 *6 00 *9 00 0 75 0 75 *2 50 4 00 *2 50 I 50 0 75 0 75 I 00 8 D. VAN NOSTRAxND CO.'S SHORT TITLE CATALOG Crosskey, L. R., and Thaw, J. Advanced Perspective.... 8vo, 2 00 Culley, J. L. Theory of Arches i6mo, 0 75 Gushing, H. C, Jr., and Harrison, N. Central Station Management ... ^a 00 Dadourian, H. M. Analytical Mechanics i2mo, *3 00 Graphic Statics 8vo, 075 Danby, A. Natural Rock Asphalts and Bitumens » 8vo, *2 50 Darling, E. R. Inorganic Chemical Synonyms iimo, i 00 Davenport, C. The Book 8vo, 2 00 Davey, N. The Gas Turbine 8vo, *4 00 Davies, F. H. Electric Power and Traction 8vo, *2 00 — — Foundations and Machinery Fixing i6mo, i 00 Deerr, N. Sugar Cane 8vo, 10 00 De la Coux, H. The Industrial Uses of Water 8vo, 5 00 Del Mar, W. A. Electric Power Conductors 8vo, *2 00 Denny, G. A. Deep-level Mines of the Rand 4to, *io 00 De Roos, J. D. C. Linkages lömo, o 75 Derr, W. L. Block Signal Operation Oblong i2mo, *i 50 Desaint, A. Three Hundred Shades and How to Mix Them 8vo, *9 00 De Varona, A. Sewer Gases . . i6mo, 075 Devey, R. G. Mill and Factory Wiring i2mo, i 00 Dichmann, Carl. Basic Open- Hearth Steel Process i2mo, 4 00 Dieterich, K. Analysis of Resins, Balsams, and Gum Resins. .. .8vo, *3 50 Dilworth, E. C. Steel Railway Bridges 4to. '''4 00 Dinger, Lieut. H. C. Care and Operation of Naval Machinery. .. i2mo, *3 00 Dixon, D. B. Machinist's and Steam Engineer's Practical Calculator. lömo, morocco, i 25 Dommett, W. E. Motor Car Mechanism i2mo, '""'a 00 Dorr, B. F. The Surveyor's Guide and Pocket Table-book. löino, morocco, 2 00 Draper, C. H. Heat and the Principles of Thermo-Dynamics. .i2mo, 2 25 Draper, E. G. Navigating the Ship i2mo, 2 00 Dubbel, H. High Power Gas Engines 8vo, *5 00 Dumesny, P., and Noyer, J. Wood Products, Distillates, and Extracts. 8vo, *5 00 Duncan, W. G., and Penman, D. The Electrical Equipment of Collieries. 8vo, *5 00 Dunkley, W. G. Design of Machine Elements. Two volumes. .8vo,each, 2 00 Dunstan, A. E., and Thole, F. B. T. Textbook of Practical Chemistry. i2mo, *i 40 Durham, H. W. Saws . Svo, 2 50 Duthie, A. L. Decorative Glass Processes Svo, 2 50 Dwight, H. B. Transmission Line Formulas Svo, *2 00 Dyke, A. L. Dyke's Automobile and Gasoline Engine Encyclopedia, Svo, 5 00 Dyson, S. S. A Manual of Chemical Plant. 12 parts. .. .410, paper, 7 50 Dyson, S. S., and Clarkson, S. S. Chemical Woiks 8vo, ^9 00 Eccles, W. H. Wireless Telegraphy and Telephony i2mo, *8 80 2 50 0 50 ^I 50 5 00 4 00 9 00 4 00 6 25 5 00 '3 00 I 00 ^i 25 7 50 »^a 00 5 00 5 00 "i 50 •=1 00 •=2 00 D. VAN NOSTRAND CO.'S SKORT TITLE CATALOG Eck, J. Light, Radiation and Illumination 8vo, Eddy, L. C. Laboratory Manual of Alternating Currents lamo, Edelman, P. Inventions and Patents i2mo, Edgcumbe, K. Industrial Electrical Measuring Instruments 8vo, Edler, R. Switches and Switchgear Svo, Eissler, M. The Metallurgy of Gold 8vo, The Metallurgy of Silver 8vo, The Metallurgy of Argentiferous Lead Svo, A Handbook on Modern Explosives 8vo, Ekin, T. C. Water Pipe and Sewage Discharge Diagrams folio, Electric Light Carbons, Manufacture of Svo, Eliot, C. W., and Storer, F. H. Compendious Manual of Qualitative Chemical Analysis i2mo, Ellis, C. Hydrogenation of Oils Svo, ultraviolet Light, Its Applications in Chemical Arts izmo, (In Press) ■ and Meigs, J. V. Gasolene and Other Motor Fuels.. (In Press.) Ellis, G. Modern Technical Drawing 8vo, Ennis, Wm. D. Linseed Oil and Other Seed Oils 8vo, ■ Applied Thermodynamics 8vo, Flying Machines To-day i2mo, Vapors for Heat Engines i2mo, Ermen, W. F. A. Materials Used in Sizing 8vo, Erwin, M. The universe and the Atom i2mo (Reprinting.) Ewing, A. J. Magnetic Induction in Iron 8vo, 5 00 Fairchild, J. F. Graphical Compass Conversion Chart and Tables... o 50 Fairie, J. Notes on Lead Ores izmo, *o 50 Notes on Pottery Clays i2mo, *2 00 Fairley, W., and Andre, Geo. J. Ventilation of Coal Mines. .. .i6mo, o 75 Fairweather, W. C. Foreign and Colonial Patent Laws 8vo, *3 00 Falk, K. G. Chemical Reactions: Their Theory and Mechanism. (In Press.) Fanning, J. T. Hydraulic and Water-supply Engineering 8vo, *S 00 Farnsworth, P. V. Shop Mathematics. . . i2mo (In Press.) Fay, I. W. The Coal-tar Dyes 8vo, 5 oa Fernbach, R. L. Glue and Gelatine , 8vo, *3 00 Findlaj^ A. The Treasures of Coal Tar i2mo, 2 00 Firth, J. B. Practical Physical Chemistry i2mo, i 25 Fischer, E. The Preparation of Organic Compounds T2mo, i 50 Fisher, H. K. C, and Darby, W. C. Submarine Cable Testing. . ,8vo, 4 00 Fleischmann, W. The Book of the Dairy Svo, 4 50 Fleming, J. A. The Alternate-current Transformer. Two Volumes. 8vo. Vol. I. The Induction of Electric Currents *6 50 Vol, II. The Utilization of Induced Currents 6 50 Propagation of Electric Currents 8vo, 3 50 A Handbook for the Electrical Laboratory and Testing Room. Two Volumes 8vo, each, *6 5° Fleur}', P. Preparation and Uses of White Zinc Paints 8vo, 3 00 Flynn, P. J. Flow of Water i2mo, o 75 Hydraulic Tables r6mo, o 75 0 75 0 75 15 oo *i oo 2 oo I 00 I oo 5 oo *5 oo 10 oo lO D. Va.N NOSTRAND CO.'S SHORT TITLE CATALOG Foster, H. A. Electrical Engineers' Pocket-book. {Seventh Edition.) i2mo, leatlier, 5 00 Engineering Valuation of Public Utilities and Factories ....*.. .8vo, *3 00 Fowle, F. F. Overhead Transmission Line Crossings i2mo, *i 50 The Solution of Alternating Current Problems 8vo. (In Press.) Fox, W. G. Transition Curves i6mo, o 75 Fox, W., and Thomas, C. W. Practical Course in Mechanical Draw- ing , . . . . i2mo, I 25 Foye, J. C. Chemical Problems lömo, Handbook of Mineralogy i6mo, Francis, J. B. Lowell Hydraulic Experiments 4to, Franzen, H. Exercises in Gas Analysis i2mo, Fräser, E. S., and Jones, R. B. Motor Vehicles and Their Motors, 8vo, fabrikoid, Freudemacher, P. W. Electric Mining Installations .lamo. Friend, J. N. The Chemistry of Linseed Oil i2mo, Fritsch, J. Manufacture of Chemical Manures 8vo, Frye, A. I. Civil Engineers' Pocket-book i2mo, leather, Fuller, G. W. Investigations into the Purification of the Ohio River. 4to, Fumell, J. Paints, Colors, Oils, and Varnishes 8vo. Gant, L. W. Elements of Electric Traction .8vo, *2 50 Garcia, A. J. R. V. Spanish-English Railway Terms 8vo, 3 00 Gardner, H. A. Paint Researches, and Their Practical Applications, 8vo, *5 00 Garforth, W. E. Rules for Recovering Coal Mines after Explosions and Fires i2mo, leather, Garrard, C. C. Electric Switch and Controlling Gear 8vo, Gaudard, J. Foundations i6mo, Gear, H. B., and Williams, P. F. Electric Central Station Distribution Systems 8vo, Geerligs, H. Ci P. Cane Sugar and Its Manufacture 8vo, Chemical Control m Cane Sugar Factories 4to, Geikie, J. Structural and Field Geology 8vo, Mountains. Their Growth, Origin and Decay 8vo, The Antiquity of Man in Europe 8vo, Georgi, F., and Schubert, A. Sheet Metal Working 8vo, Gerhard, W. P. Sanitation, Watersupply and Sewage Disposal of Country Houses i2mo, Gas Lighting i6mo, Household Wastes i6mo, House Drainage i6mo, Sanitary Drainage of Buildings i6;mo, Gerhardi, C. W. H. Electricity Meters 8vo, Geschwind, L. Manufacture of Alum and Sulphates 8vo, Gibbings, A. H. Oil Fuel Equipment for Locomotives. 8vo. (Reprinting.) Gibbs, W. E. Lighting by Acetylene i2mo, *i 50 I 50 *6 00 0 75 *3 50 =■=6 00 5 00 4 50 4 50 *3 00 3 50 ^■2 00 0 75 0 75 0 75 0 75 *7 20 5 00 6 oo 2 50 '3 50 '2 00 'l 00 I 25 ■2 50 ^2 00 '3 00 2 50 2 00 2 50 4 50 I 00 0 75 0 75 o 75 '''2 OC D. VAN NOSTRAND CO.'S SHORT TITLE CATALOG n Gibson, A. H. Hydraulics and Its Application 8vo, Water Hammer in Hydraulic Pipe Lines i^mo, Gibson, A. H., and Ritchie, E. G. Circular Arc Bow Girder ^to, Gilbreth, F. B. Motion Study i2mo, Primer of Scientific Management lamo, Gillmore, Gen. Q. A. Roads, Streets, and Pavements i2mo, Godfrey, E. Tables for Structural Engineers i6mo, leather, Golding, H. A. The Theta-Phi Diagram i2mo, Goldschmidt, R. Alternating Current Commutator Motor 8vo, Goodchild, W. Precious Stones 8vo, Goodell, J. M. The Location, Construction and Maintenance of Roads 8vo, Goodeve, T. M. Textbook on the Steam-engine i2mo. Gore, G. Electrolytic Separation of Metals 8vo, Gould, E. S. Arithmetic of the Steam-engine i2mo, Calculus i6mo, High Masonry Dams 1 6mo, Gould, E. S. Practical Hydrostatics and Hydrostatic Formulas. .i6mo, Gratacap, L. P. A Popular Guide to Minerals 8vo, Gray, H. H. Gas-Works Products 8vo (In Prrss. ) Gray, J. Electrical Influence Machines 12 mo, 2 00 Marine Boiler Design i2mo (RcpyirJing.) Greenhill, G. Dynamics of Mechanical Flight 8vo, *2 50 Greenwood, H. C. The Industrial Gases 8vo (In Press.) Gregorius, R Mineral Waxes i2mo, Grierson, R. Some Modern Methods of Ventilation 8vo, Griffiths, A. B. A Treatise on Manures lamo (Rcprinliiu/.) Gross, E. Hops Svo, Grossman, J. Ammonia and Its Compounds i2nio, Groth, L. A. Welding and Cutting Metals by Gases or Electricity. 8vo, Grover, F. Modern Gas and Oil Engines 8vo, Grüner, A. Power-loom Weaving 8vo, Grunsky, C. E. Topographic Stadia Surveying i6mo, Guldner, H. Internal Combustion Engines (/;; Press.) Günther, C. 0. Integration 8vo, Gurden, R. L. Traverse Tables folio, half morocco, Guy, A. E. Experiments on the Flexure of Beams Svo, Haenig, A. Emery and Emery Industry Svo, Hainbach, R. Pottery Decoration i2mo, Hale, A. J. The Manufacture of Chemicals by Electrolysis. 8vo, (III Press ) Hale, W. J. Calculations of General Chemistry i2mo, Hall, C. H. Chemistry of Paints and Paint Vehicles i2mo, Hall, R. H. Governors and Governing Mechanism i2mo. Hall, W. S. Elements of the Differential and Integral Calculus. . . .8vo, Descriptive Geometry 8vo volume and a 4to atlas, Haller, G. F., and Cunningham, E. T. The Tesla Coil i2mo, Halsey, F. A. Slide Valve Gears i2rao, The Use of the Slide Rules i6mo, Worm and Spiral Gearing T5mo, 3 *3 00 00 I 00 50 2 *3 *3 2 50 00 50 00 I =^7 50 50 =^1 25 "2 3 50 50 I 50 *2 00 *2 2 4 50 75 00 *i 25 I 50 0 0 75 75 I 50 *I 50 *4 oo *2 oo I 50 *I 50 2 50 6 00 *3 00 3 50 2 00 "2 50 =i--2 00 *2 00 ( 12 D. VAN NOSTRAND CO.'S SHORT TITLE CATALOG Hancock, H. Textbook of Mechanics and Hydrostatics 8vo, Hardy, E. Elementary Principles of Graphic Statics i2mo, Haring, H. Engineering Law. Vol. I. Law of Contract 8vo, Harper, J. H. Hydraulic Tables on the Flow of Water i6nio, Harris, S. M. Practical Topographical Surveying {In Press.) Harrow, B. Eminent Chemists of Our Times: Their Lives and Work. (In Press.) Haskins, C. H. The Galvanometer and Its Uses i6mo, Hatt, J. A. H. The Colorist square i2mo, Hausbrand, E. Drying by Means of Air and Steam lamo, Evaporating, Condensing and Cooling Apparatus , . .8vo, Hausmann, E. Telegraph Engineering .8vo, Hausner, A. Manufacture of Preserved Foods and Sweetmeats. .. .Svo, Hawkesworth, J. Graphical Handbook for Reinforced Concrete Design. 4to, Hay, A. Continuous Current Engineering Svo, Hayes, H. V. Public Utilities, Their Cost New and Depreciation. . .8vo, Public utilities, Their Fair Present Value and Return 8vo, Heath, F. H. Chemistry of Photography 8vo. {In Press.) Heather, H. J. S. Electrical Engineering 8vo, 450 Heaviside, 0. Electromagnetic Theory. Vols. I and II....8vo, each, {Reprinting.) Vol. Ill 8vo {Reprinting.)- Heck, R. C. H. The Steam Engine and the Turbine 8vo, 4 50 Steam-Engine and Other Steam Motors. Two Volumes. Vol. I. Thermodynamics and the Mechanics 8vo, 4 50 Vol. II. Form, Construction, and Working 8vo, 5 50 Notes on Elementary Kinematics 8vo, boards, *i 00 Graphics of Machine Forces 8vo, boards, *i 00 Heermann, P. Dyers' Materials i2mo, 3 00 Hellot, Macquer and D'Apligny. Art of Dyeing Wool, Silk and Cotton. Svo, *2 00 Hering, C, and Getman, F. H. Standard Tables of Electro-Chemical Equivalents i2mo, *2 00 Hering, D. W. Essentials of Physics for College Students 8vo, 2 25 Herington, C. F. Powdered Coal as Fuel Svo, 3 00 Herrmann, G. The Graphical Statics of Mechanism i2mo, 2 00 Herzfeld, J. Testing of Yarns and Textile Fabrics Svo. {New Edition in Preparation.) Hildenbrand, B. W. Cable-Making i6mo, o 75 Hilditch, T. P. A Concise History of Chemistry i2mo, *i 50 Hill, M. J. M. The Theory of Proportion 8vo, Hillhouse, P. A. Ship Stability and Trim 8vo, Hiroi, I. Plate Girder Construction i6mo, > Statically-Indeterminate Stresses i2mo, Hirshfeld, C. F. Engineering Thermodynamics i6mo. Hoar, A. The Submarine Torpedo Boat i2mo, Hobart, H. M. Heavy Electrical Engineering 8vo, Design of Static Transformers i2mo, Electricity 8vo, Electric Trains 8vo, Electric Propulsion of Ships 8vo, *2 50 4 50 0 75 2 50 0 75 *2 00 H=4 50 2 50 *2 00 *2 50 =1=2 50 D. VAN xNOSTRAND CO.'S SHORT TITLE CATALOG 13 Hobart, J. F. Hard Soldering, Soft Soldering and Brazing lamo, 1 25 Hobbs, W. R. P. The Arithmetic of Electrical Measurements. .. .i2mo, o 75 Hoff, J. N. Paint and Varnish Facts and Formulas i2mo, i 50 Hole, W. The Distribution of Gas 8vo, *8 50 Hopkins, N. M. Model Engines and Small Boats i2mo, i 25 The Outlook for Research and Invention i2mo. 2 00 Hopkinson, J., Shoolbred, J. N., and Day, R. E. Dynamic Electricity. i6mo. o 7"^ Horner, J. Practical Ironfounding 8vo, *2 00 G«ar Cutting, in Theory and Practice 8vo {Reprinting.) Houghton, C. E. The Elements of Mechanics of Materials lamo, 2 50 Houstoun, R. A. Studies in Light Production i2mo, 2 00 Hovenden, F. Practical Mathematics for Young Engineers i2mo, *i 50 Howe, G. Mathematics for the Practical Man i2mo, i 50 Howorth, J. Repairing and Riveting Glass, China and Earthenware. 8vo, paper, *o 50 Hoyt, W. E. Chemistry by Experimentation 8vo, *o 70 Hubbard, E. The Utilization of Wood- waste Bvo, *2 50 Hübner, J. Bleaching and Dyeing of Vegetable and Fibrous Materials. 8vo {Reprinting.) Hudson, O. F. Iron and Steel 8vo, ^ ^o Humphreys, A. C. The Business Features of Engineering Practice . 8vo, 2 50 Hunter, A. Bridge Work 8vo. {In Press.) Hurst, G. H. Handbook of the Theory of Color 8vo, *3 5° Dictionary of Chemicals and Raw Products 8vo, *5 00 Lubricating Oils, Fats and Greases 8vo, *5 00 Soapa 8vD, *6 00 Hurst, G. H., and Simmons, W. H. Textile Soaps and Oils 8vo, 3 50 Hurst, H. E., and Lattey, R. T. Text-book of Physics Svo, *3 00 Also published in three parts. Part I. Dynamics and Heat i So Part IL Sound and Light i 5» Part III. Magnetism and Electricity *i 5» Hutchinson, R. W., Jr. Long Distance Electric Power Transmission. i2mo, *3 00 Hutchinson, R. W., Jr., and Thomas, W. A. Electricity in Mining. i2mo, {In Press.) Hyde, E. W. Skew Arches i6mo, Eyde, F. S. Solvents, Oils, Gums, Waxes 8vo, Induction Coils i6mo. Ingham, A. E. Gearing. A practical treatise 8vo, Ingle, H. Manual of Agricultural Chemistry Bvo {In Press.' Inness, C. H. Problems in Machine Design i2mo, CeQtrifogal Pumps i2mo, The Fan i2mo, Jacob, A., and Gould, E. S. On the Designing and Construction of Storage Reservoirs i6mo. o 75 Jacobs, F. B. Cam Design and Manufacture ( In Press.) 0 75 *2 00 0 V.S *2 50 *3 00 *3 00 ^^4 00 H D. VAN NOSTRAND CO.'S SHORT TITLE CATALOG James, F. D. Controllers for Electric Motors 8vo. ^ oo Jehl, F. Manufacture of Carbons 8vo. .■; oo Jennings, A. S. Commercial Paints and Painting 8vo. 2 so Jennison, F. H. The Manufacture of Lake Pigments. .Bvo {In Press.) Jepson, G. Cams and the Principles of their Construction 8vo, *i 50 Mechanical Drawing 8vo {In Preparation.) Jervis-Smith, F. J. Dynamometers Bvo. 4. 00 Jockin, W. Arithmetic of the Gold and Silversmith i2mo, *i 00 Johnson, C. H., and Earle, R. P. Practical Tests for the Electrical Laboratory ( /;; Press.) Johnson, J. H. Arc Lamps and Accessory Apparatus i2mo, o 75 Johnson, T. M. Ship Wiring and Fitting i2mo (Reprinting.)', Johnston, J. F. W., and Cameron, C. Elements of Agricultural Chemistry and Geology - izmo, 2 60 Joly, J. Radioactivity and Geology i2mo {Reprinting.) Jones, H. C. Electrical Nature of Matter and Radioactivity i2mo, Nature of Solution Bvo, . New Era in Chemistry lamo, Jones, J. H. Tinplate Industry Bvo, Jones, M. W. Testing Raw Materials Used in Paint lamo, Jordan, L. C. Practical Railway Spiral i2mo, leather, Jiiptner, H. F. V. Siderdogy: The Science of Iron Bvo, -i-2 00 *3 50 *2 00 *3 00 *2 50 *I *5 50 00 . 0 75 u 00 *2 50 2 00 *0 50 *I 50 Kapp, G. Alternate Current Machinery i6mo, Kapper, F. Overhead Transmission Lines 4to, Keim, A. W. Prevention of Dampness in Buildings .Bvo, Keller, S. S., and Knox, W. E. Analytical Geometry and Calculus... Kemble, W. T,, and tJnderhill, C. R. The Periodic Law and the Hydrogen Spectrum 8vo, paper, Kemp, J. F. Handbook of Rocks 8vo, Kennedy, A. B. W., and Thurston, R. H. Kinematics of Machinery. i6mo, o 75 Kennedy, A. B. W., Unwin, W. C, and Idell, F. E. Compressed Air. i6mo, ICennedy, R. Flying Machines; Practice and Design i2mo, Principles of Aeroplane Construction Bvo, Kent. W. Strength of Materials i6mo, Kershaw, J. B. C. Fuel, Water and Gas Analysis. .Bvo (In Press.) Electrometallurgy Bvo, Electro-Thermal Methods of Iron and Steel Production. .. .Bvo, Kinzbrunner, C. Continuous Current Armatures Bvo, — — Testing of Alternating Current Machines 8vo, Kinzer, H., and Walter, K. Theory and Practice of Damask Weaving, Bvo, 4 00 Kirkaldy, A.. W., and Evans, A. D. History and Economics of Transport < Bvo, *3 00 Kirkbride, J. Engraving for Illustration Bvo, *i 00 Kirschke, A. Gas and Oil Engines.... ; lamo, *i 50 0 75 2 50 '2 00 0 75 2 50 "3 00 1 50 •2 00 D VAN XOSTRAND CO.'S SHORT TITLE CATALOG 15 Klein, J. F. Design of a High-speed Steam-engine 8vo, *5 00 Physical Significance of Entropy 8vo *i so Klingenberg, G. Large Electric Power Stations *.".".."!. .'....4to, *5 00 Knight, R.-Adm. A. M. Modern Seamanship 8vo -6 50 Pocket Edition lamo,' fabrikoid,' 3 00 Knott, C. G., and Mackay, J. S. Practical Mathematics 8vo, 2 50 Knox, J. Physico-Chemical Calculations i2mo i "50 Fixation of Atmospheric Nitrogen /.."..'. i2mo' 00 o 50 00 6 00 GO Koester, F. Steam-Electric Power Plants 4to, *5 00 Hydroelectric Developments and Engineering. . 4to *s 00 Koller, T. The utilization of Waste Products '. ' " ' 8vo *? 00 Cosmetics o ' ^r^ Koppe, S. W. Glycerine ■■■■■.■.■.■.■.■.■.'.■.■.■.■.■.■.;; i^mo' 4 .0 Kozmin, P. A. Flour Milling *"■■ gvo' 7^0 Krauch, C. Chemical Reagents '.'.'.'.'..". Svo' 7 00 Kremann, R. Application of the Physico-Chemical Theory to Tech- nical Process and Manufacturing Methods Svo ■? 00 Kretchmar, K. Yarn and Warp Sizing '. ' ^gvo] *5 00 Laffargue, A. Attack in Trench Warfare i6mo Lanier, E. V. Elementary Manual of the Steam Engine. ." i2mo' ^ uu Lambert, T. Lead and Its Compounds ' ' ' ' gvo 4 en Bone Products and Manures 8 ' *^ Z Lamborn, L. L. Cottonseed Products 8vo^ 4 00 Modem Soaps, Candles, and Glycerin Svo' *7 en Lamprecht, R. Recovery Work After Pit Fires '.'.V.V.Svo' 5 00 Lanchester, F. W. Aerial Flight. Two Volumes. Svo. Vol. I. Aerodynamics Vol. II. Aerodonetics :(:^ Lanchester, F. W. The Flying Machine '.'.'.'.'. Vvo *^ 00 Industrial Engineering: Present and Post-War Outlook. ■.Viamo! i 00 Lange, K. R. By-Products of Coal-Gas Manufacture lamo 2 50 La Rue, B. F. Swing Bridges g ' Lassar-Cohn, Dr. Modern Scientific Chemistry .' .' .' .' .' .' .' ' ' ' .' .' .' [ .' .' izmo,' 2 25 Latimer, L. H, Field, C. J., and Howell, J. W. Incandescent Electric Lighting Latta, M. N. Handbook of American Gas ■Engineering' Practice S^^o' s 00 American Producer Gas Practice 4to ' *6 00 Laws, B. C. Stability and Equilibrium of Floating Bodies 8vo' 4 50 Lawson, W. R. British Railways. A Financial and Commercial Survey _ LeSv t' T '^'"'T'^^ Machinery: :::;;:;:;: ...no' '(k^pM^l) ' °° Danger Angle '^"°' 5 00 Le Doux, M. Ice-Making Machin^::'.[:'.::'.[:,[:::[:\ llmo I yl Leeds C. C. Mechanical Drawing for Trade Schools oblong 4to: 2 25 Mechanical Drawing for High and Vocational Schools... So 1 ?o Principles of Engineering Drawing svo (In Press.) Lefevre, L. Architectural Pottery.. ,i.„ Lehner, S. Ink Manufacture .. i^°' l °° Lemstrom, S. Electricity in Agriculture and Horticulture: ! .'i i;.' Svo! *i 50 Letts, E. A. Fundamental Problems in Chemistry svo' *2 00 Le Van, W. B. Steam-Engine Indicator , ■.■.■.Viemo; 075 3 oc- '5 00 =^=o 60 I 50 '15 00 ■15 00 ^15 00 *2 50 15 00 *2 50 15 00 10 00 *I 00 4 50 4 50 2 00 2 50 0 75 I 5» 16 D. VAN NÜSTRANÜ lO.'S SHORT "ITLE CATALOG Lewes, V. B. Liquid and Gaseous Fuels .8vo, Carbonization of Coal 8vo, Lewis Automatic Machine Rifle; Operation of .i6mo, Licks, H. E. Recreations in Mathematics i2mo, Lieber, B. F. Lieber's Five Letter American Telegraphic Code .8vo, '■ • ■ Spanish Edition 8vo, '■ ■ French Edition 8vo, ^ Terminal Index 8vo, Lieber's Appendix folio, ^ Handy Tables 4*0, Bankers and Stockbrokers' Code and Merchants and Shippers' Blank Tables 8vo, =< 100,000,000 Combination Code 8vo, '^ Livermore, V. P., and Williams, J. How to Become a Competent Motor- man i2mo, Livingstone, R. Design and Construction of Commutators 3vo, Mechanical Design and Construction of Generators 8vo, Lloyd, S. L. Fertilizer Materials i2mo, Lockwood, T. D. Electricity; Magnetism, and Electro-telegraph .... Svo, — Electrical Measurement and the Galvanometer lamo, Lodge, O. J. Elementary Mechanics i2mo, Loewenstein, L. C, and Crissey, C. P. Centrifugal Pumps 5 00 Lomax, J. W. Cotton Spinning i2mo, i 50 Lord, R. T. Decorative and Fancy Fabrics Svo, *3 50 Loring, A. E. A Handbook of the Electromagnetic Telegraph. . .i6mo, o 75 Lowy, A. Organic Type Formulas o 10 Lubschez, B. J» Perspective i2mo, *i 50 Lucke, C. E. Gas Engine Design 8vo, *?. 90 — — Power Plants: Design, Efficiency, and Power Costs. 2 vols. (In Preparation.) Luckiesh, M. Color and Its Application , . .8vo, Light and Shade and Their Applications 8vo, Visual Illusions (In Preparation.) Lunge, G. Coal-tar and Ammonia. Three Volumes 8vo, Technical Gas Analysis 8vo, Manufacture of Sulphuric Acid and Alkali. Four Volumes. . . Svo, Vol. I. Sulphuric Acid. In three parts (Reprinting.) Vol. I. Supplement 8vo ( Reprinting.) Vol. IL Salt Cake, Hydrochloric Acid and Leblanc Soda. In two parts (In Pri'ss!) Vol. III. Ammonia Soda (In Press.) Vol. IV. Electrolytic Methods (In Press.) Technical Chemists' Handbook lamo, leather, *4 00 Technical Methods of Chemical Analysis. Vol. I. In two parts ., .8vo, *i5 00 Vol. II. In two parts , 8vo, *i8 00 Vol. III. In two parts 8vo, *i8 00 The set (3 vols.) complete *5o 00 Luquer, L. M. Minerals in Rock Sections 8vo, *i 5°> 3 50 3 CO ^25 CO *4 50 D. VAN NOSTRAND CO.'S SHORT TITLE CATALOG 17 MacBride, J. D. A Handbook of Practical Shipbuilding, 121710, fabrikoid, 2 00 Macewen, H. A. Food Inspection 8vo, *2 50 Mackenzie, N. F. Notes on Irrigation Works 8vo, *2 50 Mackie, J. How to Make a Woolen Mill Pay 8vo, *2 00 Maguire, Wm. R. Domestic Sanitary Drainage and Plumbing . . . .8vo, 4 00 Malcolm, H. W. Submarine Telegraph Cable 9 00 Malinovzsky, A. Analysis of Ceramic Materials ' and Methods of Calculation {Jn Press.) Mallet, A. Compound Engines i6mo, Mansfield, A. N, Electro-magnets i6mo, o 75 Marks, E, C. R. Construction of Cranes and Lifting Machinery. lamo, *2 75 Manufacture of Iron and Steel Tubes lamo, 2 50 Mechanical Engineering Materials i2mo, *i 50 Marks, G. C. Hydraulic Power Engineering 8vo, 4 50 Marlow, T. G. Drying Machinery and Practice. .. .8vo (Reprinting.) Marsh, C. F. Concise Treatise on Reinforced Concrete Svo, *2 50 Reinforced Concrete Compression Member Diagram. Mounted on Cloth Boards *i . 50 Marsh, C. F., and Dunn, W. Manual of Reinforced Concrete and Con- crete Block Construction i6mo, cloth, 2 00 Marshall, W. J., and Sankey, H. R. Gas Engines 8vo, 2 00 Martin, G. Triumphs and Wonders of Modern Chemistry 8vo, *3 00 Modern Chemistry and Its Wonders 8vo, *3 00 Martin, N. Properties and Design of Reinforced Concrete 8vo, i 50 Martin, W. D. Hints to Engineers i2mo, 2 00 Massie, W. W., and Underhill, C. R. Wireless Telegraphy and Telephony. i2mo, *i 00 Mathot, R. E. Internal Combustion Engines 8vo, 5 00 Maurice, W. Electric Blasting Apparatus and Explosives .Svo, *3 50 Shot Firer's Guide Svo, *i 50 Maxwell, F. Sulphitation in White Sugar Manufacture i2mo, 4 00 Maxwell, J. C. Matter and Motion i6mo, o 75 Maxwell, W. H., and Brown, J. T. Encyclopedia of Municipal and Sani- tary Engineering .4to, *io 00 Mayer, A. M. Lecture Notes on Physics Svo, 2 00 McCracken, E. M., and Sampson, C. H. Course in Pattern Making. (/;; Press.) McCullough, E. Practical Surveying i2mo, 2 50 McCullough, R. S. Mechanical Theory of Heat Svo, 3 50 McGibbon, W. C. Indicator Diagrams for Marine Engineers 8vo, "3 50 Marine Engineers' Drawing Book oblong 4to, *2 50 McGibbon, W. C. Marine Engineers Pocketbook i2mo, *4 50 Mcintosh, J. G. Technology of Sugar 8vo, *6 00 • Industrial Alcohol 8vo, *3 50 Manufacture of Varnishes and Kindred Industries. Three Volumes. Svo. Vol. I. Oil Crushing, Refininq^ and Boiling 7 00 Vol. II. Varnish Materials and Oil Varnish Malting. (Reprinting.) Vol. III. Spirit Varnishes and Materials (Reprinting.) McKay, C. W. Fundamental Principles of the Telephone Business. 8vo. (In Press.) 2 oo 2 00 1 00 2 00 2 50 2 50 '^2 00 2 25 *I 00 *I 00 *3 00 18 D. VAN NOSTRAND CO.'S SHORT TITLE CATALOG McKillop, M., and McKillop, A. D. Efficiency Methods i2mo, i 50 lIcKnight, J. D., and Brown, A. W. Marine Multitubular Boilers.... *2 50 McMaster, J. B. Bridge and Tunnel Centres i6mo, o 75 McMechen, F. L. Tests for Ores, Minerals and Metals i2mo, *i co McNair, Jas. B. Citrus By-Products {In pi\\';s. ) Heade, A. Modern Gas Works Practice 8vo, *8 50 Melick, C. W. Dairy Laboratory Guide lamo, *i 25 •^Mentor." Self-Instruction for Students in Gas Supply. i2mo. Elementary 2 50 Advanced 2 50 Self -Instruction for Students in Gas Engineering. i2mo. Elementarj' Advanced Merivale, J. H. Notes and Formulae for Mining Students i2mo, Merritt, Wm. H. Field Testing for Gold and Silver. .. .i6rao, leather, Mertens. Tactics and Technique cf River Crossings 8vo, Mierzinski, S. Waterproofing of Fabrics 8vo, Wiessner, B. F. Radio Dynamics i2mo, Miller, G. A. Determinants. i6mo, Miller, W. J. Introduction to Historical Geology i2mo, Milroy, M. E. W. Home Lace-making i2mo, Mills, C. N. Elementary Mechanics for Engineers 8vo, Mitchell, C. A. Mineral and Aerated Waters 8vo, Mitchell, C. A., and Prideaux, R. M. Fibres Used in Textile and Allied Industries 8vo, 3 50 Mitchell, C. F., and G. A. Building Construction and Drawing. i2mo. Elementary Course 2 00 Advanced Course 3 00 Monckton, C. C. F. Radiotelegraphy 8vo, 2 00 Monteverde, R. D. Vest Pocket Glossary of English-Spanish, Spanish- English Technical Terms 64mo, leather, *i 00 Montgomery, J. H. Electric Wiring Specifications iSmo, *i 00 Moore, E. C. S. New Tables for the Complete Solution of Ganguillet and Kutter's Formula 8vo, Moore, Harold. Liquid Fuel for Internal Combustion Engines ... 8vo, Morecroft, J. H., and Hehre, F. W. Short Course in Electrical Testing. 8vo, Morgan, A. P. Wireless Telegraph Apparatus for Amateurs i2mo, Morrell, R. S., and Waele, A. E. Rubber, Resins, Paints and Var- nishes 8vo (In Press. ) Moses, A. J. The Characters of Crystals 8vo, and Parsons, C. L. Elements of Mineralogy 8vo, Most, S. a. Elements of Gas Engine Design i6mo, -The Lay-out of Corliss Valve Gears i6mo, Mulford, A. C. Boundaries and Landmarks i2mo, Mulford, A. C. Boundaries and Landmarks i2mo, Munby, A. E. Chemistry and Physics of Building Materials. .. .8vo, Murphy, J. G. Practical Mining i6mo, Murray, B. M. Chemical Reagents 8vo ( In Press. ) Murray, J. A. Soils and Manures 8vo, 2 00 Nasmith, J. The Student's Cotton Spinning 8vo, 4 50 Recent Cotton Mill Construction lamo, 3 00 '^6 00 5 00 I 75 *l 50 *2 00 *3 50 0 75 0 75 *i 00 I 00 2 50 I 00 19 I 50 ♦2 00 3 50 7 50 •■•I 00 10 00 I 00 7 5c 0 75 15 00 6 50 -r-j 00 4 00 0 75 4 00 2 00 *3 00 D. VAN NÜSTRAND CO.'S SHORT TITLE CATALOG Weave, G. B., and Heilbron, I, M. Identification of Organic Compounds. i2mo, Neilson, R. M. Aeroplane Patents 8vo, JNerz, F. Searchlights 8vo (Rcpriiüing.) Newbigin, M. I., and Flett, J. S. James Geikie, the Man and the Geologist 8vo, Newbiging, T. Handbook for Gas Engineers and Managers 8vo, Newell, F. H., and Drayer, C. E. Engineering as a Career. .i2mo, cloth, Nicol, G. Ship Construction and Calculations 8vo, Nipher, F. E. Theory of Magnetic Measurements i2mo, Nisbet, H. Grammar of Textile Design 8vo, Nolan, H. The Telescope i6mo, Norie, J. W. Epitome of Navigation (2 Vols.) octavo, A Complete Set of Nautical Tables with Explanations of Their Use octavo, North, H. B. Laboratory Experiments in General Chemistry lamo, O'Connor, H. The Gas Engineer's Pocketbook lamo, leather, Ohm, G. S., and Lockwood, T. D. Galvanic Circuit i6mo, Olsen, J. C. Text-book of Quantitative Chemical Analysis 8vo, Orrasby, M. T. M. Surveying i2mo, Oudin, M. A. Standard Polyphase Apparatus and Systems 8vo, Pakes, W, C. C, and Nankivell, A. T. The Science of Hygiene . .8vo, *i 75 Palaz, A. Industrial Photometry 8vo, 4 00 Palmer, A. R. Electrical Experiments lamo, o 75 Magnetic Measurements and Experiments i2mo, o 75 Pamely, C. Colüery Manager's Handbook Svo, *io 00 Parker, P. A. M. The Control of Water 8vo, 600 Parr, G. D. A. Electrical Engineering Measuring Instruments. .. .8vo, *3 50 Parry, E. J. Chemistry of Essential Oils and Artificial Perfumes. Vol. I. Monographs on Essential Oils 9 oc Vol. II. Constituents of Essential Oils, Analysis 7 00 Foods and Drugs. Two Volumes. Vol. I. The Analysis of Food and Drugs 8vo, Vol.11. The Sale of Food and Drugs Acts 8vo, and Coste, J. H. Chemistry of Pigments Svo, Parry, L. Notes on Alloys 8vo, Metalliferous Wastes 8vo, Analysis of Ashes and Alloys 8vo, Parry, L. A. Risk and Dangers of Various Occupations Svo, Parshall, H. F., and Eobart, H. M. Electric Railway Engineering. 4to, Parsons, J. L. Land Drainage Svo, Parsons, S. J. Malleable Cast Iron 8vo {Reprinting.) Partington, J. R. Higher Mathematics for Chemical Students. .i2mo, Textbook of Thermodynamics 8vo, The Alkali Industry 8vo, Patchell, W. H. Electric Power in Mines Svo, pT.ei-'^nr,, G. V/. L. V/i'-'n? Ca'cr^at'ons lamo, Electri'^ M'-"e S'o-"gU''"or I'l-^tallat^'ons i2mo, Patter-or, D. The Color Printing of Carpet Yarns 8vo, f'oloT Ifl'^t'^hir.or 0-1 Textiles 8vo, Textile Color Mixir.g 8vo, 9 50 3 50 "5 00 ■'■3 50 ■^2. 50 •^2 50 "3 50 7 50 ••'I 50 2 50 "4 00 3 00 *4 CO *2 r '^ ''■I 5^ "3 50 "3 50 ■■3 50 20 D. VAN NOSTRAND CO.'S SHORT TITLE CATALOG Paulding, C. P. Condensation of Steam in Covered and Bare Pipes. .8vo, Transmission of Heat through Cold-storage Insulation i2mo, Payne, D. W. Iron Founders' Manual 8vo, Peddie, R. A. Engineering and Metallurgical Books 12 mo, Peirce, B. System of Analytic Mechanics 4to, Linear Associative Algebra 4to, Perkin, F. M., and Jaggers, E. M. Elementary Chemistry. ... .lamo, Perrin, J. Atoms 8vo, Perrine, F. A. C. Conductors for Electrical Distribution 8vo, Petit, G. White Lead and Zinc White Paints. 8vo, Petit, R. How to Build an Aeroplane 8vo, Pettit, Lieut. J. S. Graphic Processes i6mo, Philbrick, P. H. Beams and Girders i6mo, Phin, J. Seven Follies of Science 12 mo, Pickw^orth, C. N. Logarithms for Beginners i2mo, boards, The Slide Rule i2mo, Pilcher, R. B. The Profession of Chemistry izmo (In Press.) Pilcher, R. B,, and Butler-Jones, F. What Industry Owes to Chemical Science lamo, Plattner's Manual of Blow-pipe Analysis. Eighth Edition, revised. Sv«, Ply mp ton, G. W. The Aneroid Barometer i6mo, How to Become an Engineer i6mo, Van Nostrand's Table Book i6mo. Pochet, M. L. Steam Injectors i6mo. Pocket Logarithms to Four Places i6mo, i6mo, leather, Polleyn, F. Dressings and Finishings for Textile Fabrics 8vo, Pope, F. G. Organic Chemistry i2mo. Pope, F. L. Modern Practice of the Electric Telegraph .8vo, Popplewell, W. C. Prevention of Smoke Svo, Strength of Materials 8vo, Porritt, B. D. The Chemistry of Rubber i2mo, Porter, J. R. Helicopter Flying Machine . . , i2mo. Potts, H. E. Chemistry of the Rubber Industry Svo, Practical Compounding of Oils, Tallows and Grease Bvo, Pratt, A. E. The Iron Industry 8vo (In Press.) • The Steel Industry Bvo (In Press.) Pratt, Jas. A. Elementary Machine Shop Practice (In Press.) Pratt, K. Boiler Draught i2mo, Prelini, C. Earth and Rock Excavation Bvo, Graphical Determination of Earth Slopes Svo, Tunneling. New Edition 8vo, Dredging. A Practical Treatise 8vo, Prescott, A. B., and Johnson, 0. C. Qualitative Chemical Analysis. .8vo, Prescott, A. B., and Sullivan, E. C. First Book in Qualitative Chemistry. Prideaux, E. B. R. Problems in Physical Chemistry 8vo, The Theory and Use of Indicators 8vo, Prince, G. T. Flow of Water i2mo, Pull, E. Modern Steam Boilers 8vo, PuUen, W. W. F. Application of Graphic Methods to the Design of i2mo, Structures i2mo, Injectors: Theory, Construction and Working. i2mo, Indicator Diagrams Svo, Engine Testing 8vo, +2 00 *I 00 4 00 *I SO 10 00 2 50 I CO *2 50 *3 50 *2 GO I 50 0 75 *I SO I 00 I 50 I 50 4 00 0 75 0 75. 0 75 0 75 0 75 I OO' ^3 50 2 50 I 50 *3 50. H 50 I 00 I 50 2 50 0 50 *I 2S' *3 00 *2 00 *3 00 *3 00 4 00 *2 oa 5 00 *2 00 5 oa *r 50 3 00 -■=2 GO- 3 00 ''5 5^ 0 75 o 75 ' 0 75 6 00 *4 00 *5 00 4 00 7 50 6 00 6 00 3 50 3 50 2 oo 2 50 ^■5 00 *i 50 D. VAN NOSTRAND CO.'S SHORT TITLE CATALOG 21 Purday, E. F. P. The Diesel Engine Design 8vo (/;; Press.) Putsch, A. Gas and Coal-dust Firing 8vo, *2 50 Rafter, G. W. Mechanics of Ventilation i6mo, Potable Water i6mo, Treatment of Septic Sewage i6mo, and Baker, M. N. Sewage Disposal in the United States. .. .4to, Raikes, H. P, Sewage Disposal Works 8vo, Randau, P. Enamels and Enamelling 8vo, Rankine, W. J. M., .and Bamber, E. F. A Mechanical Text-book. .Svo, ■ Civil Engineering Svo. Machinery and Millwork Svo, The Steam-engine and Other Prime Movers Svo, Rankine, W. J. M., and Bamber, E. F. A Mechanical Text-book. . . .8vo, Raphael, F. C. Localization of Faults in Electric Light and Power Mains. Svo, Easch, E. Electric Arc Phenom.ena 8vo, Rathbone, R. L. B. Simple Jewellery Svo, Eausenberger, F. The Theory of the Recoil Guns 8vo, Rautenstrauch, W» Notes on the Elements of Machine Design. 8 vo, boards, Hautenstrauch, W., and Williams, J. T. Machine Drafting and Empirical Design. Part I. Machine Drafting 8vo, i 50 Part II. Empirical Design {In Preparation.) Raymond, E. B. Alternating Current Engineering i2mo, *2 50 Rayner, H. Silk Throwing and Waste Silk Spinning 8vo, üecipes for the Color, Paint, Varnish, Oil, Soap and Drysaltery Trades, Svo, ^'s 00 Recipes for Flint Glass Making i2mo, *5 00 Redfern, J. B., and Savin, J. Bells, Telephones i6mo, o 75 Redgrove, H. S. Experimental Mensuration i2mo, i 50 Reed, S. Turbines Applied to Marine Propulsion *5 00 Reed's Engineers' Handbook 8vo, ^g 00 Key to the Nineteenth Edition of Reed's Engineers' Handbook. .Svo, 4 00 Useful Hints to Sea-going Engineers i2mo, 3 00 Reid, E. E. Introduction to Research in Organic Chemistry, (In Press.) Reinhardt, C. W. Lettering for Draftsmen, Engineers, and Students. oblong 4to, boards, i 00 Reinhardt, C. W. The Technic of Mechanical Drafting, oblong, 4to, boards, *i 00 Reiser, F. Hardening and Tempering of Steel i2mo, 2 50 Reiser, N. Faults in the Manufacture of Woolen Goods Svo, 2 50 Spinning and Weaving Calculations 8vo, ""5 00 Renwick, W. G. Marble and Marble Working. .. .Svo (Reprinting.) Reuleaux, F. The Constructor 4to, 4 00 Rey, Jean. The Range of Electric Searchlight Projectors Svo, (Reprinti)ig.) Reynolds, C, and Idell, F. E. Triple Expansion Engines i6mo, o 75 Rhead, G. F. Simple Structural Woodwork. i2mo, *i 25 Rhead, G. W. British Pottery Marks Svo, 3 50 Rhodes, H. J. Art of Lithography Svo, 5 00 Rice, J. M., and Johnson, W. W. A New Method of Obtaining the Differ- ential of Functions • ramo, 0 50 24 D. VAN NOSTRAND CO.'S SHORT TITLE CATALOG Sherriff, F. F. Oil Merchants' Manual and Oil Trade Ready Reckoner, 8vo, 3 50 Shields, J. E. Notes on Engineering Construction. . . i2mo, i 50 Shreve, S. H. Strength of Bridges and Roofs 8vo, 3 50 Shunk, W. F. The Field Engineer i2mo, fabrikoid, 2 50 Silverman, A., and Harvey, A. W. Laboratory Directions and Study Questions in Inorganic Chemistry 4to, loose leaf, 2 00 Simmons, W. H. Fats, Waxes and Essential Oils. .8vo (In Press.) Simmons, W. H., and Appleton, H. A. Handbook of Soap Manufacture, 8/0, *4 00 Simmons, W. H., and Mitchell, C. A. Edible Fats and Oils Svo, *3 50 Simpson, G. The Naval Constructor i2mo, fabrikoid, *5 00 Simpson, W. Foundations Svo. {In Press.) Sinclair, A. Development of the Locomotive Engine. . . 8vo, half leather, 5 00 Sindall, R. W. Manufacture of Paper Svo (Reprinting.) Sindall, R. W., and Bacon, W. N. The Testing of Wood Pulp 8vo, iRepri)}ting.) Wood and Cellulose .8vo (/;; Press.) Sloane, T. O'C. Elementary Electrical Calculations i2mo, Smallwood, J. C. Mechanical Laboratory Methods. ...lamo, fabrikoid, Smith, C. A. M. Handbook of Testing, MATERIALS 8vo, Smith, C. A. M., and Warren, A. G. New Steam Tables 8vo, Smith, C. F. Practical Alternating Currents ard Testing 8vo, Practical Testing of Dynamos and Motors Svo, Smith, F. E. Handbook of General Instruction for Mechanics . . . i2mo, Smith, G. C. Trinitrotoluenes and Mono- and Dinitrotoluenes, Their Manufacture and Properties i2mo, Smith, H. G. Minerals and the Microscope i2mo, Smith, J. C. Manufacture of Paint Svo, Smith, R. H. Principles of Machine Work i2mo, Advanced Machine Work i2mo, Smithy W. Chemistry of Hat Manufacturing i2mo, Snell, F. D. Calorimetric Analysis i2mo (In Press.) Snow, W. G., and Nolan, T. Ventilation of Buildings i6mo, o 75 Soddy, F. Radioactivity 8vo (Reprinting.) Solomon, M. Electric Lamps Svo, 2 00 Somerscales, A. N. Mechanics for Marine Engineers lamo, 2 50 Mechanical and Marine Engineering Science Svo, *5 00 Sothern, J. W. The Marine Steam Turbine 870, *i2 50 Verbal Notes and Sketches for Marine Engineers Svo, *i2 50 Marine Engine Indicator Cards Svo, 4 50 Sothern, J. W., and Sothern, R. M. Simple Problems in Marine Engineering Design i2mo, Souster, E. G. W. Design of Factory and Industrial Buildings.. .Svo, Southcombe, J. E. Chemistry of the Oil Industries Svo, Soxhlet, D. H. Dyeing and Staining Marble Svo, Spangenburg, L. Fatigue of Metals i6mo, Specht, G. J., Hardy, A. S., McMaster, J. B., and Walling. Topographical Surveying i5mo, Spencer, A. S. Design of Steel -Framed Sheds Svo, Spiegel, L. Chemical Constitution and Physiological Action. .. .i2mo, *2 00 3 00 *2 50 *I 25 *3 50 -3 00 I 50 2 00 2 GO *5 00 *3 00 *3 50 3 00 4 00 3 50 2 50 0 75 0 75 3 50 I 25 D. VAN NOSTRAND CO.'S SHORT TITLE CATALOG -D Sprague, E, H. Hydraulics ismo, Elements of Graphic Statics 8vo, — — Stability of Masonry i?mo, Elementary Mathematics for Engineers i2mo, Stability of Arches :?.mo, Strength of Structural Elements lamo, ^ Moving Loads by Influence Lines and Other Methods lamo, Stahl, A. W. Transmission of Power i6mo, Stahl, A. W., and Woods, A. T. Elementary Mechanism i2mo, Standage, H. C. Leatherworkerg' Manual 8vo, Sealing Waxes, Wafers, and Other Adhesives 8vo, Agglutinants of All Kinds for All Purposes lamo, Stanley, H. Practical Applied Physics. (In Press.) Stansbie, J. H. Iron and Steel Svo, Steadman, F. M. Unit Photography i2mo, Stecher, G. E. Cork. Its Origin and Industrial Uses lamo, Steinheil, A., and Voit, E. Applied Optics^ Vols. I. and II. Svo, Each, • Two Volumes Set, Steinman, D. B. Suspension Bridges and Cantilevers. (Science Series No. 127.) Melan's Steel Arches and Suspension Bridges Svo, Stevens, A. B. Arithmetric of Pharmacy i2mo, Stevens, E. J. Field Telephones and Telegraphs Stevens, H. P. Paper Mill Chemist i6mo, Stevens, J. S. Theory of Measurements lamo, Stevenson, J. L. Blast-Furnace Calculations i2mo, leather, Stewart, G. Modern Steam Traps i2mo. Stiles, A. Tables for Field Engineers i2mo, Stodola, A. Steam Turbines Svo, Stone, E. W. Elements of Radiotelegraphy i2mo, fabrikoid, Stone, H. The Timbers of Commerce Svo, Stopes, M. The Study of Plant Life Svo, Sudborough, J. J., and James, T.C. Practical Organic Chemistry .lamo, Suf fling, E. R. Treatise on the Art of Glass Painting Svo, Sullivan, T. V., and Underwood, N. Testing and Valuation of Build- ing and Engineering Materials (/;; Press.) Sutherland, D. A. The Petroleum Industry Svo (In Press.) Svenson, C. L. Handbook on Piping Svo, 4 00 Essentials of Drafting Svo, i 75 Mechanical and Machine Drawing and Design (In Press.) Swan, K. Patents, Designs and Trade Marks Svo, 2 00 Swinburne, J., Wordingham, C, H., and Martin, T. C. Electric Currents. i6mo, o 75 Swoope, C. W. Lessons in Practical Electricity i2mo, *2 00 Tailfer, L. Bleaching Linen and Cotton Yarn and Fabrics Svo, 7 00 Tate, J. S. Surcharged and Different Forms of Retaining-walls. .i6mo, o 75 Taylor, F. N. Small Water Supplies izmo, *2 50 Masonry in Civil Engineering 8vo, *2 50 Taylor, W. T. Electric Cable Transmission and Calculation. ( 7» Press.) Calculation of Electric Conductors 4to (In Press.) Templeton, W. Practical Mechanic's Workshop Companion. i2mo, morocco, 2 00 Tenney, E. H. Test Methods for Steam Power Plants i2mo, 3 00 Terry, H. L. India Rubber and its Manufacture. .Svo (Reprinting.) 2 CO 3 CO 2 00 2 00 2 00 2 00 2 00 *2 00 "5 50 "^2 50 3 50 2 50 "2 00 I 00 5 00 9 00 0 75 -^3 00 I 50 I 20 4 00 '''i 25 2 50 -i-j 75 I 00 5 00 2 50 4 00 2 00 2 50 '''3 50 26 D- VAN NOSTRAND CO.'S SHORT TITLE CATALOG Thayer, H. R. Structural Design. 8vo. Vol. I. Elements of Structural Design a 50 Vol. II. Design of Simple Structures 450 Vol. III. Design of Advanced Structures . . . ., {In Preparation.) ■ — — Foundations and Masonry {In Preparation.) Thiess, J. B., and Joy, G. A. Toll Telephone Practice 8vo, Thorn, C, and Jones, W. H. Telegraphic Connections.. . .oblong, i2mo, Thomas, C. W. Paper-makers' Handbook {In Press.) Thomas, J. B. Strength of Ships 8vo, Thomas, Roht. G. Applied Calculus i2mo, Thompson, A. B. Oil Fields of Russia 4to, Oil Field Development 10 00 Thompson, S. P. Dynamo Electric Machines i6mo, Thompson, W. P. Handbook of Patent Law of All Countries i6mo, Thomson, G. Modern Sanitary Engineering i2mo, Thomson, G. S. Milk and Cream Testing i2mo, Modem Sanitary Engineering, House Drainage, etc 8vo, Thornley, T. Cotton Combing Machines , 8vo, Cotton Waste 8vo, Cotton Spinning. Svo. First Year Second Year Third Year Thurso, J. W. Modem Turbine Practice 8vo, Tidy, C. Meymott. Treatment of Sewage i6mo, rilmans, J. Water Purification and Sewage Disposal 8vo, Tiiuiey, W. H. Gold-mining Machinery 8vo, Titherley, A. W. Laboratory Course of Organic Chemistry 8vo, Tizard, H. T. Indicators (In Press.) Toch, M. Chemistry and Technology of Paints 8vo, Materials for Permanent Painting i2mo. Tod, J., and McGibbon, W. C. Marine Engineers' Board of Trade Examinations 8vo, Todd, J., and Whall, W. B. Practical Seamanship 8vo, Townsend, F. Alternating Current Engineering 8vo, boards, *o 7S Townsend, J. S. Ionization of Gases by Collision 8vo, *i 25 Transactions of the American Institute of Chemical Engineers, 8vo. Vol. I. to XL, 1908-1918 8vo, each, Traverse Tables • i6mo, Treiber, E. Foundry Machinery i2mo, Trinks, W. Governors and Governing of Prime Movers 8vo, Trinks, W., and Housum, C. Shaft Governors i6mo, Trowbridge, W. P. Turbine Wheels i6mo, Tucker, J. H. A Manual of Sugar Analysis 8vo, Tumbull, Jr., J., and Robinson, S. W. A Treatise on the Compound Steam-engine i6mo, Turner, H. Worsted Spinners' Handbook i2mo, Turrill, S. M. Elementary Course in Perspective i2mo, Twyford, H. B. Purchasing 8vo, Storing, Its Economic Aspects and Proper Methods Svo, *3 SO I 50 2 50 3 00 10 00 10 00 0 75 I 50 '"S 00 *2 25 *3 00 '^'S 50 *3 50 Vj 50 "'3 *2 50 50 *4 00 0 75 2 50 *3 00 *2 00 4 50 *2 00 *2 00 12 00 6 00 0 75 2 00 3 50 0 75 0 75 3 SO 0 75 ^-3 00 »I 25 '3 00 3 50 D. VAN NOSTRAND CO.'S SHORT TITLE CATALOG 27 underbill, C. R. Solenoids, Electromagnets and Electromagnetic Wind- ings i2mo, 3 00 underwood, N., and Sullivan, T. V. Chemistry and Technology of Printing Inks 8vo, *3 00 Urquhart, J, W. Electro-plating lamo, 2 00 Electrotjrping ... i2mo, 2 00 üsborne, P. O. G. Design of Simple Steel Bridges 8vo, *4 00 Vacher, F. Food Inspector's Handbook lamo, Van Nostrand's Chemical Annual. Fourth issue igiS.fabrikoid, lamo, *3 00 Year Book of Mechanical Engineering Data (hi Press.) Van Wagenen, T. F. Manual of Hydraulic Mining i6mo, i 00 Vega, Baron Von. Logarithmic Tables 8vo, 2 50 Vincent, C. Ammonia and its Compounds. Trans, by M. J. Salter. 8vo, *2 50 Vincent, C. Ammonia and its Compounds 8vo, 2 50 Virgin, R. Z. Coal Mine Management (In Press.) Volk, C. Haulage and Winding Appliances 8vo, *4 00 Von Georgievics, G. Chemical Technology of Textile Fibres. .. .8vo, Chemistry of Dyestuffs 8vo, (Nen' Edition in Preparation.) Vose, G. L, Graphic Method for Solving Certain Questions in Arithmetic and Algebra i6mo, o 75 Vosmaer, A. Ozone 8vo, *2 50 Wabner, R. Ventilation in Mines 8vo, 5 00 vVadmore, T. M. Elementary Chemical Theory i2mo, *i 50 Wagner, E. Preserving Fruits, Vegetables, and Meat i2mo, ="2 50 Wae;ner, H. E., and Edwards, H. W. Railway Engineering Estimates. (In Press.) Wagner, J. B. Seasoning of Wood 8vo, 3 00 Waldram, P. J. Principles of Structural Mechanics i2mo, 4 00 Walker, F. Dynamo Building i6mo, o 75 Walker, J. Organic Chemistry for Students of Medicine 8vo, 4 oc Walker, S. F. Steam Boilers, Engines and Turbines 8vo, 3 00 -Refrigeration, Heating and Ventilation on Shipboard i2mo, *2 50 — — Electricity in Mining 8vo, *4 50 Electric Wiring and Fitting 8vo, 2 50 Wallis-Tayler, A. J. Bearings and Lubrication 8vo, *i 50 Aerial or Wire Ropeways Svo (Reprinting.) Preservation of Wood 8vo, 4 00 Refrigeration, Cold Storage and Ice Making . 8vo, 5 50 Sugar Machinery i2mo, 3 00 Walsh, J. J. Chemistry and Physics of Mining and Mine Ventilation, i2mo, 2 50 Wanklyn, J. A. Water Analysis i2mo, 2 00 Wansbrough, W. D. The A B C of the Differential Calculus. . . .i2mo, *2 50 Slide Valves i2mo, *2 00 Waring, Jr., G. E. Sanitary Conditions i6mo, o 75 Sewerage and Land Drainage *6 00 Modern Methods of Sewage Disposal i2mo, 2 00 How to Drain a House i2mo, i 25 28 D. VAN NOSTRAND CO.'S SHORT TITLE CATALOG Warnes, A. R. Coal Tar Distillation 8vo, *s OO' Warren, F. D. Handbook on Reinforced Concrete i2mo, *2 50 Watkins, A. Photography 8ve, 3 50. Watson, E. P. Small Engines and Boilers i2mo, i 25 Watt, A. Electro-plating and Electro-refining of Metals 8vo, 5 00 Electro-metallurgy i2mo, i OO' Paper-Making 8vo, 3 75 Leather Manufacture 8vo, 6 00 The Art of Soap Making Svo, 4 00 Webb, H. L. Guide to the Testing of Insulated Wires and Cables. i2mo, i 00 Wegmann, Edward. Conveyance and Distribution of Water for Water Supply 8vo, 5 00 Weisbach, J. A Manual of Theoretical Mechanics 8vo, *6 00 Weisbach, J., and Herrmann, G. Mechanics of Air Machinery. .. .8vo, ^3 75 Wells, M. B. Steel Bridge Designing 8vo, *2 50 Wells, Robt. Ornamental Confectionery i2mo, 3 00 Weston, E. B. Loss of Head Due to Friction of Water in Pipes. .i2mo, 2 00 Wheatley, 0. Ornamental Cement Work 8vo, *2. 25 Whipple, S. An Elementary and Practical Treatise on Bridge Building. 8vo, 3 00 White, C. H. Methods of Metallurgical Analysis i2mo, 250 White, G. F. Qualitative Chemical Analysis i2mo, i 4a White, G. T. Toothed Gearing i2mo, *2 00 White, H. J. Oil Tank Steamers i2ino, i 50 Whitelaw, John. Surveying Svo, 4 50 Whittaker, C. M. The Application of the Coal Tar Dyestuffs. ..8vo, 3 00 Widmer, E. J. Military Balloons 8vo, 3 00 Wilcox, R. M. Cantilever Bridges i6mo, o 75 Wilda, H. Steam Turbines i2mo, 2 00 Cranes and Hoists i2mo, 2 00 Wilkinson, H. D. Submarine Cable Laying and Repairing 8vo, {Reprinting.) Williamson, J. Surveying Svo, *3 00 Williamson, R. S. Practical Tables in Meteorology and Hypsometry, 4t0, 2 50 Wilson,- F. J., and Heilbron, I. M. Chemical Theory and Calculations. i2mo, *i 25. Wilson, J. F. Essentials of Electrical Engineering 8vo, 2 50 Wimperis, H. E. Internal Combustion Engine Svo, "^3 00 — — Application of Power to Road Transport i2mo, *i 50 Primer of Internal Combustion Engine i2mo, i 50 Winchell, N. H., and A. N. Elements of Optical Mineralogy Svo, *3 50 Winslow, A. Stadia Surveying i6mo, o 75 Wisser, Lieut. J. P. Explosive Materials i6mo, Modern Gun Cotton i6mo, o 75 Wolff, C. E. Modern Locomotive Practice 8vo, *4 20 Wood, De V. Luminif erous Aether i6mo, o 75 Wood, J. K. Chemistry of Dyeing i2mo, i OO' Worden, E. C. The Nitrocellulose Industry. Two Volumes 8vo, *io 00 Technology of Cellulose Esters. In 10 volumes. 8vo. Vol. VIII. Cellulose Acetate *5 00 D. VAN NOSTRAND CO.'S SHORT TITLE CATALOG 29 Wren, H. Organometallic Compounds of Zinc and Magnesium. .i2mo, i 00 Wiight, A. C. Analysis of Oils and Allied Substances 8vo, *3 50 Wright, A. C. Simple Method for Testing Painters' Materials. . .8vo, 2 50 Wright, F. W. Design of a Condensing Plant.. i2mo {Rcpriiitmg.) Wright, H. E. Handy Book for Brewers 8vo, *6 00 Wright, J. Testing, Fault Finding, etc., for Wiremen i5mo, 0 50 Wright, T. W. Elements of Mechanics Bvo, "2 50 Wright, T. W., and Hayford, J. F. Adjustment of Observations. . .Bvo, *3 00 Wynne, W. E., and Sparagen, W. Handbook of Engineering Mathe- matics i2mo, 2 00 Yoder, J. H., and Wharen, G. B. Locomotive Valves and Valve Gears, 8vo, *3 00 Young, J. E. Electrical Testing for Telegraph Engineers Bvo, *4 00 Young, R. B. The Banket Bvo, 3 50 Youngson. Slide Valva and Valve Gears. 8vo, 3 00 Zahner, R. Transmission of Power i6mo, Zeuner, A. Technical Thermodynamics. Two Volumes Bvo, 8 00 Zimmer, G. F. Mechanical Handling and Storing of Materials. .. .4to, 15 00 Mechanical Handling of Material and Its National Importance During and After the War 4to, 4 00 2ipser, J. Textile Raw Materials 8vo, 5 00 Zur Nedden, F. Engineering Workshop Machines and Processes. .Bvo, 2 00 are prepared to supply, either from their complete stock or at short notice, Any Technical or Scientific Book In addition to publishing a very large and varied number of Scientific and Engineering Books, D. Van Nostrand Company have on hand the largest assortment in the United States of such books issued by American and foreign publishers. All inquiries are cheerfully and care- fully answered and complete catalogs sent free on request. 25 Park Place - - - New York "-»9)100*1