THE DECENNIAL PUBLICATIONS OF THE UNIVERSITY OF CHICAGO THE DECENNIAL PUBLICATIONS ISSUED IN COMMEMORATION OP THE COMPLETION OP THE FIRST TEN YEARS OP THE UNIVERSITY'S EXISTENCE AUTHORIZED BY THE BOARD OP TRUSTEES ON THE RECOMMENDATION OP THE PRESIDENT AND SENATE EDITED BY A COMMITTEE APPOINTED BY THE SENATE EDWAED CAPPS STAKE WILLABD CUTTING EOLLIN D. SALISBURY JAMES BOWLAND ANGELL WILLIAM I. THOMAS SHAILEE MATHEWS GAEL DARLING BUCK FEEDEEIC IVES CAEPENTEE OSKAE BOLZA JULIUS 8TIEGLITZ JACQUES LOEB THESE VOLUMES ARE DEDICATED TO THE MEN AND WOMEN OP OUR TIME AND COUNTRY WHO BY WISE AND GENEROUS GIVING HAVE ENCOURAGED THE SEARCH AFTER TRUTH IN ALL DEPARTMENTS OF KNOWLEDGE DIFFUSION AND OSMOTIC PRESSURE THE ROLE OF DIFFUSION AND OSMOTIC PRESSURE IN PLANTS BY BUKTON EDWAKD LIVINGSTON OF THE DEPARTMENT OP THE DECENNIAL PUBLICATIONS SECOND SERIES VOLUME VIII CHICAGO THE UNIVERSITY OF CHICAGO PRESS 1903 c. * Copyright 1903 BY THE UNIVERSITY OF CHICAGO PREFACE WITH the ever-increasing tendency to regard an organ- ism as a complex of physical and chemical processes which may one day be analyzed and understood, there has neces- sarily gone hand in hand a tendency toward more and more accurate and quantitative investigation of the physics and chemistry of the cell itself. Among the various groups of physical and chemical phenomena that have been found to play important roles in the life-process, and which, there- fore, have been interrogated for answers to physiological questions, none has stood out within the past few years as more fundamentally important than those connected with diffusion and osmotic pressure. This field has thus far only been touched upon, and it would seem, judging from researches which have recently appeared, that the best and most far-reaching work therein is probably yet to come. The present volume will deal with the past and present of diffusion and osmotic pressure from the standpoint of plant physiology. It has a double raison d'etre. First, it was felt that there was need of some direct and not too exhaustive account of the essential physical facts and theories of the subject. The interest of the physical chemist here has lain mainly in the light which these phenomena have been able to throw upon the ultimate nature of matter and upon elec- trolytic processes. It has thus been difficult for the student of physiology who is not at the same time well versed in physical chemistry to obtain the information required for the prosecution of work in this field. Secondly, it seemed desirable to bring together in a general review the literature of this subject in its biological aspects, so that the promising and unpromising points for future research might become x PREFACE more apparent. The volume will thus naturally fall into two Parts, the first dealing with the purely physical aspect of these phenomena, and the second attempting to present in a more or less unified whole the physiological results which have so far appeared in this connection, together with their bearing upon each other and upon the vital problem as a whole. Chapter IV of Part II was presented to the Faculty of the Ogden Graduate School of Science of the University of Chicago in candidacy for the doctor's degree in 1901. The author wishes here to express his thanks to Professor C. K. Barnes, of this laboratory, and to Professor Jacques Loeb, of the Hull Physiological Laboratory, for much valu- able aid. Professor Barnes has kindly read the manuscript and has made many suggestions. The author alone is, how- ever, responsible for whatever new departures are to be found in the book. B. E. L. THE HULL BOTANICAL LABORATORY, The University of Chicago, October 1, 1902. TABLE OF CONTENTS PREFACE PART I. PHYSICAL CONSIDERATIONS INTRODUCTION - - - 1 CHAPTER I. Matter and Its States ------ 3 i. Fundamental Theories of the Nature of Matter, a) The Atomic Theory. 5) The Kinetic Theory. ii. The Three States of Matter. CHAPTER II. Diffusion and Diffusion Tension 9 i. Gases. a) Simple Gases. b) Mixed Gases, ii. Liquids. a) Simple Liquids. 6) Mixed Liquids, in. Solids. a) Simple Solids. b) Diffusion of Two Solids. CHAPTER III. Liquid Solutions 16 i. Liquids in Liquids, ii. Gases in Liquids, in. Solids in Liquids, iv. Terminology for Solutions of Differing Concentration. CHAPTER IV. lonization 23 i. Of Gases, ii. Of Solutes in Liquid Solutions. CHAPTER V. Osmotic Phenomena 25 i. Osmotic Pressure of the Solute. a) Non-electrolytes. b) Electrolytes. c) Colloids. d) Osmotic Pressure in General. xii TABLE OF CONTENTS ii. Diffusion Tension of the Solvent, in. Experimental Demonstration of Osmotic Pressure. CHAPTER VI. Measurement and Calculation of Osmotic Pressure 35 i. Measurement of Osmotic Pressure, a) Direct Method. 6) Indirect Methods. 1. Freezing-point. 2. Boiling-point. 3. Vapor Tension. ii. Calculation of Osmotic Pressure, a) Non-electrolytes. 6) Electrolytes. PART II. PHYSIOLOGICAL CONSIDERATIONS INTRODUCTION .- 47 CHAPTER I. Turgidity - - 49 i. Protoplasm and its Limiting Membranes, ii. Plasmolysis. in. The Permeability of the Protoplasmic Layers, a) Test by the Plasmolytic Method. 6) Direct Test of Penetrability. c) Absorption Test. d) Test by Toxicity. e ) Test by Accumulation. /) Test by Metabolic Processes. g) Outward Permeability. h) Variations in Permeability, iv. Action of the Protoplasmic Membrane. a) The Filter Theory. b) The Solution Theory. c ) The Chemical Theory. v. The Nature of the Osmotically Active Solutes, vi. The Maintenance of Turgidity in Spite of Permeability to Certain Solutes. vn. The Relation of Turgidity to Vital Activity. a) The Retention of Form. 6) Mechanical Support. c) Growth. d) Curvature. e) Work. vni. Summary of the Chapter. TABLE OF CONTENTS xiii CHAPTER II. Absorption and Transmission of Water - - 91 i. Absorption of Water, ii. Transmission of Water. a) Water Loss. (1) Evaporation. (2) Water Pores and Nectaries. (3) Exudation. (4) Summary of Water Loss. &) Upward Movement of Water in Trees and Other Tall Plants, in. Summary of the Chapter. CHAPTER III. Absorption and Transmission of Solutes - - 115 i. Absorption of Gases. ii. Absorption of Dissolved Solids and Liquids, in. Transmission of Solutes. CHAPTER IV. Influence of the Osmotic Pressure of the Sur- rounding Medium upon Organisms - - - 124 i. Introductory, ii. Presentation of Material. a) Influence upon Growth and Form. b) Influence upon Reproduction. c) Influence upon Irritability. (1) Changes of Irritability. (2) Osmotaxis. d) Analogy between the Effects of High Osmotic Pres- sure of the Medium and Those Produced by Other Water-Extracting Processes. in. Summary of the Chapter. INDEX 145 PART I PHYSICAL CONSIDERATIONS INTRODUCTION IN the following treatment of the physical phenomena of diffusion and osmotic pressure no attempt is made to be exhaustive. Certain aspects only of the present conceptions of these matters among most physicists and chemists1 are discussed, and every discussion is presented with the sole aim of clearing the way for the physiological discussions which are to follow. Thus, for example, the whole subject of atomic and molecular weights and their experimental determination — so important to the chemist, but not pri- marily interesting to the physiologist — is entirely omitted. Also it may be added that no attention is given to a his- torical treatment of this part of the subject, the excellent chemical treatises which are now available rendering this unnecessary. In the present Part very little is original with the author, excepting the mode of presentation. The various texts and the original papers have been drawn upon wherever it has seemed expedient. Footnotes give the names of the authors i A general confusion among younger students with regard to the way in which these conceptions take the form of theories makes it seem expedient to call atten- tion to the following points : A scientific theory does not pretend to state the truth. It may sometimes state a part of the truth, but this is not primarily its aim. Its aim is to connect the facts together in the most logical and plausible manner pos- sible, and thus to aid the further advancement of our knowledge. Its " employment has its origin in the organization of the human mind, which handles abstract truths much less easily by themselves than by the help of an illustrative image A hypothesis [or theory] can neither be proved nor disproved. It is merely a tool which is rejected when found to be no longer serviceable What the 'real' nature of matter is, is to us a matter of complete ignorance, as it is of complete indifference." (OsxwALD-WALKEB, Outlines of General Chemistry [London, 1895], pp. 5, 7.) A principle, on the other hand, does attempt to state the truth ; it is a generalization and induction from a great number of known facts. When a fact is discovered which cannot be included under a principle, then that principle falls to the ground and ceases to be a principle. What was at first a theory may at length, by the accumulation of evidence, come to be a principle. 1 INTRODUCTION to whom we are indebted for the more important points. The student of this subject will find the following standard texts very helpful: NERNST, W. Theoretical Chemistry from the Standpoint of Avo- gadro's Rule and Thermodynamics. Translated by C. S. Palmer. London, 1895. OSTWALD, W. Lehrbuch der allgemeinen Chemie, comprising : Verwandschaftslehre, Leipzig, 1887 ; Stoechiometrie, Leip- zig, 1891 ; Chemische Energie, Leipzig, 1893. Solutions. Translated by M. M. P. Muir. London, 1891. Outlines of General Chemistry. Translated by James Walker. London, 1895. Manual of Physico- Chemical Measurements. Translated by James Walker. London, 1894. The Principles of Inorganic Chemistry. Translated by Alex- ander Findlay. London, 1902. REMSEN, I. Principles of Theoretical Chemistry. New York, 1897. WALKER, J. Introduction to Physical Chemistry. London, 1899. REYCHLER, A. Outlines of Physical Chemistry. Translated by J. McCrae. London, 1899. DASTRE, M. A. "Osmose," in Trait6 de physique biologiqtie, publi^ sous la direction de MM. D'Arsonval, Gariel, Chauveau et Marey. Tome I. Paris, 1901. COHEN, ERNST. Vortrdge filr Aerzte uber physikalische Chemie. Leipzig, 1901. JONES, H. C. The Elements of Physical Chemistry. New York, 1902. WHETHAM, W. C. D. Solution and Electrolysis. Cambridge, 1895. KOHLRADSCH, F., AND HoLBORN, L. Das Leitvermogen der Elektro- lyte. Leipzig, 1898. TRAUBE, J. Physico- Chemical Methods. Translated by W. L. Hardin. Philadelphia, 1898. BLITZ, HENRY. Practical Methods for Determining Molecular Weights. Translated by Jones and King. Easton, Pa., 1899. HAMBURGER, H. J. Osmotischer Druck und lonenlehre. Wies- baden, 1902. CHAPTER I MATTER AND ITS STATES I. FUNDAMENTAL THEORIES OF THE NATURE OF MATTER a) The atomic theory. — The whole structure of modern physical science is based upon the atomic theory. This theory supposes every mass of matter to be composed of numerous ultimate, indivisible particles, called atoms, which possess a peculiar attraction for one another. Atoms differ in the amount and nature of this attractive force, those of every chemical element being in this way different from those of every other ; but all atoms of the same element, when under the same conditions, are exactly similar. Owing to their chemical attraction, atoms seldom exist free as such, but are prone to unite into groups, thus causing the neutralization of their mutual attraction. The groups so formed are called molecules. If the atoms composing the molecules of any substance art^ajike, (i. e., of the same element), the element is said to be in the molecular condition — as opposed to the atomic or nascent condition, in which the atoms are not united to one another, but exist free as such. When the atoms forming a molecule are of different chemical elements, a compound is said to be formed. The physical and chemi- cal properties of molecules are very different from those of their component atoms, and they are also very different from those of any molecules which can be formed in any other way. But all molecules which are formed of the same elements and in the same manner are exactly similar under N the same conditions. It thus appears that the smallest par- ticle of a compound which can exist and still retain the properties of that compound is the molecule ; break this up, 3 4 DIFFUSION AND OSMOTIC PRESSURE and free atoms or new groups of atoms, with new properties, will result, the original compound having been destroyed by the process of separation. Atoms may also unite in such a way that their mutual attractive forces are but partially neutralized, thus forming incomplete parts of molecules, called ions. Under certain special conditions molecules may split into two or more ions, and some of these cases of ionization or dissociation, as the process of splitting is called, have proved very important in the development of the sub- ject of osmotic pressure. In some cases an ion may consist of a single atom which has split off from some molecule. Briefly, then, acording to the atomic theory as now made use of, the nature of any mass is dependent upon that of its component particles, these particles being atoms, molecules, or ions. The same mass may contain, at the same time, all three kinds of particles. 6) The kinetic theory of matter. — According to the kinetic theory, the particles composing any mass, whatever their nature may be, are in constant motion. This necessi- tates their being considered, not as packed closely one against another to make up the mass, but as separated from one another by continuously varying spaces. The continuous motion of the particles is probably for the most part a vibratory motion. They are supposed to move in straight lines and in the same direction until a collision occurs, when they rebound according to the principle of the reflection of moving bodies. It thus becomes necessary to consider, for comparison, the average distance apart of these particles, or their average or mean free path. This has been demonstrated to be much greater than the diameter of a single particle. A rough conception of the state of affairs within a mass of matter may be obtained by comparing the mass to a cage of angry bees. The insects in such a cage fly in straight MATTER AND ITS STATES lines to and fro, striking against each other and against the walls of the cage, ever varying their distances apart, yet always remaining equally distributed throughout the cage, i. e., always keeping their average distance apart the same. Thus far nothing has been said of any restraining force to counteract in a measure the motion of the particles and keep them from flying apart indefinitely. Such a force might be roughly compared to the walls of the cage just referred to, for it is these restraining walls which prevent the indefinite enlargement of the swarm of angry insects. More accu- rately, the restraining force in the illustration is the sum of the reactions produced by the several impacts of the moving insects against the rigid walls. There is, indeed, such an active restraining force present in all masses of matter ; it is ordinarily made evident, however, only in liquids and solids. This force is the cohesion of the particles themselves. It is probably akin to gravitation, in exhibited larger bodies, and is an inverse function of the square of the average dis- tance apart of the moving particles. That is, the mutual attraction exerted by two particles decreases at the same rate as the square of their distance apart increases. It will thus be seen that this force becomes negligible at a comparatively small distance from any particle. But the particles of liquids and solids are so near to one another that their cohe- sive force is sufficient to overcome, to a certain extent, their energy of motion and to hold most of them within certain fixed limits of space. The science of thermo-dynamics rests upon another sup- position of the kinetic theory of matter, namely, that the tem- perature of any body is directly due to the kinetic energy of its vibrating particles. Since the mass of any particle remains constant, and the kinetic energy of any moving body is, at any instant, one-half the product of its mass and the square of its velocity (KE = ^M F2), it is seen that the average kinetic 6 DIFFUSION AND OSMOTIC PRESSURE energy of a particle varies with the square of its average velocity. We neglect here, as comparatively unimpor- tant, all other forms of motion which a particle may possess, such as that of rotation, and consider only its transla- tory motion. Therefore, whenever the temperature of a quantity of matter is raised by any means, the average translatory velocity of its particles is increased. Now, the force of impact of a moving body is proportional to its momentum, which is equal to the product of its mass and velocity at the time of impact. But since one particle may strike another particle at any point in its free path, here again the average velocity must be considered. Therefore, since the mass of a particle is a constant quantity, any increase in the average velocity will cause a corresponding increase in momentum, and also in the force of impact. But the force of recoil is practically equal to the force of impact, and this latter force is the repellent force which tends to separate the particles. Thus, with rising temperature the repellent force is increased, the force of cohesion is more and more nearly overcome, and the particles become more and more widely separated. Also, with the rapid decrease in the cohesive force incident upon the increase in its acting distance, a limit is soon reached beyond which the force tending to cause separation is greater than the other, and the particles fly apart indefinitely. In this condition we say the substance is a gas. If it was a liquid or solid at the lower temperature, it has now been vaporized by heat. II. THE THREE STATES OF MATTER Matter exists in three states — the gaseous, the liquid, and the solid. In gases the kinetic energy of the particles is so great that the cohesive force is entirely overcome and the particles tend ever to increase their distance apart. From this it necessarily follows that a mass of gas in a closed MATTER AND ITS STATES vessel will completely fill it, no matter if the vessel be many times the size of the original volume of gas. This is an observed fact. If such a gas is gradually cooled (i. e., allowed to do work on some other body and thus to part with some of the kinetic energy of its particles), a condition will be reached wherein the cohesive force is greater than the repellent, and the par- ticles will remain together in a definite volume. As long as the two forces involved are nearly equal, the average free path will still be relatively great, and although the particles cling together, yet they will move very freely upon one another — a condition imperfectly simulated by the component grains in a mass of sand. In this condition the substance is said to be a liquid. Here the particles move so readily upon one another that a mass of liquid still takes the form of the containing vessel, as far as that is possible without increase in volume. In this regard liquids are very differ- ent from gases. Also, on account of the freedom of motion on the part of the particles making it up, and on account of the downward pull of gravity, the free surface of a liquid is usually approximately level. There are, indeed, certain phenomena of surface tension and adhesion which make it possible for free liquid surfaces to exist in other positions than the horizontal, but the present subject does not lead to a discussion of these. It is necessary to call attention, how- ever, to the fact that, on account of the action of the cohesive force, a peculiar surface layer of particles is formed about a liquid mass, a sort of thin skin or film, which possesses con- siderable tensile strength, and which is much less easily penetrated than the internal mass. By a continuation of the process of cooling (which must ever be thought of as a process of causing the body to give up kinetic energy by doing work, such as warming another cooler body) the liquid particles may be brought still closer 8 DIFFUSION AND OSMOTIC PRESSURE together, until cohesion becomes so strong, and hence the friction of particle upon particle so great, that the free movement upon each other just described comes practically to an end. The body is now a solid and will retain its form without surrounding walls. The particles are still in violent vibration, however. It should be stated here that the ideal gas, liquid, or solid does not exist; the hardest substances show some tendency to flow like liquids, and the most fluid substances exhibit some friction of their component particles upon one another. CHAPTER II DIFFUSION AND DIFFUSION TENSION I. GASES a) Simple gases. — As has been indicated already, it is a fundamental property of all gases that they tend to fill com- pletely any vessel in which they may be inclosed. Thus, if a cubic centimeter of oxygen is measured out at ordinary temperature and at atmospheric pressure, and is then passed into a sealed vacuum chamber, it will completely fill the chamber, no matter how large the latter may be. This process of expansion is called diffusion. Of course, in dif- fusing, the particles of which the gas is composed become distributed throughout a greater space, and hence the gas becomes less dense. This property is often stated as follows : "The particles of gases tend to separate indefinitely." Because of this tendency to expand, an outward pressure, called gas pressure, is exerted by a gas upon the walls of any chamber in which it may be confined. Gas pressure is supposed to be caused by the continuous bombardment of the walls of the inclosing vessel by the vibrating gas par- ticles. If a gas be inclosed in a chamber with elastic walls, the size of the chamber will depend upon the number of particles of gas present (i. e., its concentration) and upon the kinetic energy of the particles themselves (i. e., its tem- perature). Thus, for any temperature and amount of gas, the distension of such a chamber will cease when the inward pressure, due to the resilience of the walls and to the pres- sure of the surrounding atmosphere (unless the chamber be in a vacuum), becomes equal to the outward pressure, due to the gas. 10 DIFFUSION AND OSMOTIC PRESSURE For a given amount of gas the pressure is constant at a constant temperature. But change in temperature means simply change in the kinetic energy of the particles. There- fore a rise in temperature must cause a corresponding increase in gas pressure, and a fall a corresponding decrease. Keep- ing the pressure constant, a rise in temperature produces an increase in volume, and vice versa. It has been found experimentally that the volume of a given mass of gas under constant external pressure varies with its absolute temperature (273°+ the given temperature Centigrade). This is the principle of Gay-Lussac, sometimes called that of Charles. But if, in the elastic chamber mentioned above, the tem- perature be kept constant and the resiliency of the walls be increased, thus increasing the external pressure on the gas, the volume will be decreased. As this occurs, how- ever, the gas will increase in density, and continually more particles will strike unit area of the wall in unit time. Thus the internal pressure upon the bag will also be increased, until at length another state of equilibrium will be reached, wherein the external and internal pressures are again equal. But during the readjustment the volume of the gas has decreased. As long as the temperature (i. e., the kinetic energy of the particles) is constant, an increase in external pressure produces a decrease in volume, and a decrease in external pressure an increase in volume. Experimentally it is demonstrated that the volume of a given mass of gas at a constant temperature varies inversely as the external pressure to which it is subjected. This is the principle of Boyle. Still another principle has been discovered for gases. If the volume and temperature both remain constant, and if the number of particles is increased (i. e., the concen- tration), the pressure will be correspondingly increased. It DIFFUSION AND DIFFUSION TENSION 11 is obvious from the theoretical consideration already pre- sented that this must be true. In this case the kinetic energy of the particles is not altered, but their number has been increased, hence the increase in pressure. Also, for a given concentration and temperature, all gases exhibit the same pressure. This is called the principle of Avo- gadro. It is usually stated in a somewhat different way, namely: Equal volumes of gases, at equal temperature and pressure, contain the same number of particles. This principle holds rigorously true only for gases whose con- centration is rather low. As a gas approaches the liquid state, the principle of Avogadro, and also those of Boyle and Gray-Lussac, have to be modified. They apply only to a theoretically perfect gas. b) Mixed gases. — In a mixture of several gases each gas practically exerts its own pressure independently of the others. Thus the total pressure of a mixture of gases in a chamber is the sum of the pressures which would be exhib- ited were the gases separated and each put into a chamber of the same size as the first one, the temperature of course remaining constant. The pressures which would be thus shown are called partial pressures, and the above fact may be stated more directly, by use of this term, as follows: The total pressure of a gas mixture is practically the sum of the partial pressures of its component gases. As a gas nears the liquid state, this principle also breaks down in part, it too applying rigorously only to perfect gases. Also, if two gases be brought together so as to form two horizontal strata in a chamber, diffusion of each gas will take place just as completely as if the other gas were not present. Particles of the lower gas will pass up from the lower stratum until that gas is equally distributed through- out the chamber. Downward diffusion of the upper gas will occur simultaneously, and the result of the two processes 12 DIFFUSION AND OSMOTIC PRESSURE will be a uniform mixture of the two gases. If this pro- cess of diffusion is obstructed by a wall placed between them, the pressures of both gases will of course be exhibited independently upon the opposite sides of this wall. n. LIQUIDS a) Simple liquids. — When a liquid is heated, the kinetic energy of its particles is increased, until at length the cohe- sive force which held them together is overcome ; then they fly off from the main mass and tend ever to increase their distance apart. This is the process of vaporization by heat. As long as the temperature remains high enough, such mat- ter will remain in the gaseous state. Also many substances which are usually liquids can be vaporized at ordinary tem- peratures. Water, alcohol, and ether are examples of this. This process, however, is a slow one. It is explained theoretically in this way: Although the majority of the liquid particles cannot break away from the main mass at ordinary temperatures, yet some of them, which reach the surface with greater kinetic energy than the others, do suc- ceed in breaking through the firmer surface layer (see p. 7), and so escape as gas particles. If the chamber above the liquid be a closed one, so that the evaporated liquid cannot escape, evaporation soon apparently ceases. If some of the liquid particles come against the surface layer with sufficient force to pass through it, it is reasonable to suppose that, after escap- ing into the chamber above, some of them may again pass through this film in the opposite direction, and so re-enter the liquid. Here they come under the influence of the force of cohesion, which holds the liquid particles together, and, since they are unable to break forth at once, they remain in the liquid state. The number which thus re-enter the liquid will gradually increase as the pressure of the vapor (i. e., the number of vapor particles, for the temperature is sup- DIFFUSION AND DIFFUSION TENSION 13 posedly constant) increases. Thus, an equilibrium will be established sooner or later, wherein the number of particles escaping from the liquid in unit time will be just equaled by the number re-entering it. That is, evaporation is just equaled by the opposite process, condensation. This is the condition when evaporation apparently ceases. The gas pressure with which the liquid particles escape is termed vapor tension. And when evaporation has apparently ceased, the gas pressure of the vapor in the space above the liquid is equal to the vapor tension which the particles exhibit in leaving the liquid surface. We have thus a means for measuring the vapor tension of any liquid. If the temperature rises, the vapor tension rises corre- spondingly, following the principles of gases. If the external pressure upon the supernatant mass of vapor be increased, its gas pressure becomes greater than the vapor tension of the liquid, and condensation surpasses evaporation, thus decreasing the number of vapor particles — and hence the pressure due to them — until equality of tension and pressure is restored. If two such chambers in which the supernatant vapor is at different pressures be connected above the level of the liquid, the substance will distil over and condense in the chamber which has the lower pressure. This will continue until the two pressures have been equal- ized by the resulting change in the relative volumes occupied respectively by the two masses of liquid, and by the diffusion of the vapor particles themselves. 6) Mixed liquids. — If two different, equally miscible liquids are brought into contact with each other so as to form two horizontal strata, diffusion will take place in both directions, just as in the corresponding case with gases, but much more slowly on account of the friction and interfer- ence of the particles: and there will result a uniform mix- ture in which both kinds of particles are equally distributed 14 DIFFUSION AND OSMOTIC PRESSURE throughout. This tendency of one liquid to diffuse into another may be termed diffusion tension; it corresponds to the vapor tension exhibited by an evaporating gas. Diffu- sion in liquids is sometimes distinguished from that in gases by the use of the term "hydro-diffusion" to denote the former. They are, however, essentially the same thing. In the case just cited, each liquid develops a diffusion tension independ- ently of the other. Of course, above such a mixture of liquids there will lie (if the chamber allow it) a stratum of gas mixture in which each of the two gases has its own vapor tension, just as though the other gas were not present. III. SOLIDS a) Simple solids. — Continuously raising the temperature of a solid may result in liquefying it and then in vaporizing the liquid thus formed ; or vaporization may take place immediately, without the intervention of the liquid phase at all. In either case the particles break away from the solid mass and become more widely separated. With the process of liquefaction, however, we have no concern. When vaporization of a solid takes place directly, it is called sublimation. Grum camphor, naphthalene, and ice below the temperature of melting exhibit this phenomenon. If the vapor particles are prevented from escaping, an equi- librium between vapor and solid is ultimately reached, at which sublimation apparently ceases. At such a point sub- limation is just equaled by condensation. The whole process is analogous to that of evaporation from free liquid surfaces. The pressure of the vapor surrounding a solid mass of the same substance, when equilibrium is reached, may be termed, as in liquids, vapor tension. b) Diffusion of two solids. — If two solid masses of differ- ent substances are brought together with their adjacent faces in close contact, there can be demonstrated, in some DIFFUSION AND DIFFUSION TENSION 15 cases at least,1 a diffusion of the substances into each other. The process goes on with extreme slowness, however, and the details need not be stated here. 1W. SPRING, "Ueber die chemische Einwirkung der KOrper in festem Zustande," Zeitschr.f. physik. Chem., Vol. II (1888), pp. 536-8; IDEM, "Ueber das Vorkommen gewisser fur den Flilssigkeits- oder Gaszustand charakteristischen Eigenschaften bei festen Metallen," ibid., Vol. XV (1894), pp. 65-78; W. ROBERTS-AUSTEN, " On the Diffusion of Gold in Solid Lead at the Ordinary Temperature," Proceed. Roy. Soc. London, Vol. LXVII (1901), pp. 101-5. CHAPTER III LIQUID SOLUTIONS " SOLUTIONS are homogeneous mixtures — mixtures which allow no separation of their components by mechanical means. The ability of gases to form such mixtures is un- limited, that of liquids is limited."1 Solid solutions also exist, but have not yet been shown to play any part in physiology; therefore they need not be considered here. Gas mixtures have already been discussed. There remains, then, only the subject of liquid solutions — a very important subject in the study of physiology. I. SOLUTIONS OF LIQUIDS IN LIQUIDS Not nearly all liquids are readily miscible to form solu- tions. Many are nearly — perhaps quite — insoluble in one another. Again, many liquids are mutually soluble in all proportions (e. g., water and alcohol) ; others are so only within certain limits. When a mixture of two liquids is considered as a solu- tion, the liquid which preponderates is called the solvent and the other the solute. If such a solution were brought into contact with a mass of the pure solvent, diffusion of the solute would take place into the pure solvent until the solute were uniformly distributed throughout both layers. At the same time the pure solvent would diffuse into the solution. Of course, the interchange of particles between two such layers would not cease when uniformity of constitution had been attained throughout; it would still go on, but would cease to be apparent, having become simply the continuous motion of the particles composing the uniform mixture. At i OSTWALD-WAL.KER, Outlines of General Chemistry (London, 1895), p. 117. 16 LIQUID SOLUTIONS 17 the beginning of such a process of mixing, however, a defi- nite diffusion tension exists and can be demonstrated — a dif- fusion tension produced on the one hand by the solute, and on the other by the solvent. These diffusion tensions are identical in their nature with those spoken of in the last chapter. They increase in amount with rise in temperature, and, in case there are several solutes, each one has its own diffusion tension. These facts are found to be fundamental in the consideration of osmotic pressure. Above any liquid mixture contained in a closed jar which it does not fill, there will be a gas mixture of the vapors of the solvent and of the several solutes. Each body will have its own vapor pressure, and the total pressure of the gas mixture will be the sum of its partial pressures. II. SOLUTIONS OF GASES IN LIQUIDS Gas solutes behave in the same manner as that just de- scribed for liquid solutes. The amount of gas going into solution, when a mass of it is brought into contact with a mass of liquid solvent, increases with the temperature and pressure. Diffusion pressures of solvent and solute are developed here also, and are constant for a given tempera- ture ; they also vary with the absolute temperature. There may be several gaseous solutes in the same solution, and in this case each develops its own diffusion tension in the sol- vent. Above such a solution there will be a gas mixture of the vapor of the solvent and of the several solutes. Inter- change of particles will go on continually between the gas solution above and the liquid solution below, but will not be apparent for reasons similar to those expressed above for liquid solutes. Also, if a solution containing a gas solute be brought into contact with a mass of the pure solvent, dif- fusion will take place of both solvent and solute, each devel- oping its own diffusion tension in its own direction, just as 18 DIFFUSION AND OSMOTIC PRESSURE in the corresponding case with a liquid solute. Equilibrium and an apparent stoppage of diffusion will be brought about when diffusion is equal in both directions. III. SOLUTIONS OF SOLIDS IN LIQUIDS If a crystal of sugar or salt be put into the water, it dis- solves. This process of dissolving consists in the flying off of particles into the water, just as the process of vaporiza- tion of a mass of naphthalene consists in the flying off of particles into the air. After the particles of the dissolved substance (solute) are once free from the solid mass, they behave in an entirely different manner from that which characterized them before. While they were in the crystal they clung together by cohesion. Now they tend to sepa- rate as much as possible within the limits of the solvent.1 They may or may not pass the surface of the solvent and enter the air as a gas, but within the solvent they continue to diffuse until they are uniformly distributed. Diffusion of the solute in its solvent takes place much more slowly than does gas diffusion, but in the end it is just as com- plete. Thus it is evident that within the volume occu- iT. GEAHAM, "Ueber die Diffusion von Flussigkeiten," Liebigs Ann., Vol. LXXVII (1851), pp. 56-89 and 129-60; see also ibid., Vol. LXXX (1851), pp. 197-201. The following references may serve to put the reader into contact with the literature of diffusion: T. GKAHAM, "Supplementary Observations on the Diffusion of Liquids," Phil. Trans. Roy. Soc. London, 1850, pp. 805-36; A. FICK, " Ueber Dif- fusion," Pogg. Ann., Vol. XCIV (1855), pp. 59-86; T. GRAHAM, " Anwendung der Dif- fusion der Flussigkeiten zur Analyse," Liebigs Ann., Vol. CXXI (1862), pp. 1-77; F. VoiGTiANDER, " Ueber die Diffusion in Agargallerte," Zeitschr. f. physik. Chem., Vol. Ill (1884), pp. 316-35; J. D. R. SCHEFFER, " Untersuchungen fiber die Diffusion einiger organischen und anorganischen Verbindungen," I, Ber. d. deutsch. chem. Gesellsch., Vol. XV (1882), pp. 788-801 ; II, ibid., Vol. XVI (1883), pp. 1903-17 ; W. NERNST, " Zur Einetik der in LOsung befindlichen Korper," Zeitschr. f. physik. Chem., Vol. II (1888), pp. 613-37; J. D. R. SCHEFFER, "Untersuchungen fiber die Diffusion wasse- riger Lflsungen," ibid., pp. 390-404; S. ARRHENIUS, "Untersuchungen fiber Diffusion von in Wasser gelosten Stoffen," ibid., Vol. X (1892), pp. 51-95; L. LIEBERMANN UND S. BURGARSZKY, " Beitrage zur Theorie der wasserigen LOsungen von Salzgemi- schen," ibid., Vol. XII (1893), pp. 188-95; A. NACCARI, " Sulla pressione osmotica," Atti della Beale Accad. del Lincei, Ser. 5, Rendiconti, Classe di Scienze fisiche, matemat. e naturali, Vol. II (1893), 1 Semestre, pp. 238-9, and 2 Semestre, pp. 136-8; R. HOBER, " Ueber Concentrationsanderungen bei der Diffusion zweier geloster Stoffe gegen einander," Pfliigers Archiv, Vol. LXXIV (1S99), pp. 225-45. LIQUID SOLUTIONS 19 pied by the solution, dissolved particles exhibit at least one of the fundamental properties of gas particles, namely, that of indefinite diffusion. It will be gathered from what was said under the preceding headings that the same is true for liquid and gas solutes. As in gases and in solutions with liquid and gas solutes, this tendency of the solute to diffuse (diffusion tension) may be measured. It is found that, for the same temperature and volume, the same number of particles of different solutes gives always the same diffusion tension. Thus the solute in such a solution exhibits another principle of gases, namely, that of Avogadro. This, too, is true for liquid and gas solutes. The principle does not hold rigorously for very concentrated solutions. There is developed here also a dif- fusion tension on the part of the solvent, which varies with temperature just as does that of the solute. If two solutions containing different concentrations of the same solute in the same solvent are brought into direct con- tact, it is found that diffusion of solvent and solute will at length equalize the concentrations of the two solutions, so that the solute particles will at last be equally distributed throughout the combined volume. Therefore diffusion of solute particles must be more rapid from the stronger to the weaker of the two solutions than in the opposite direction. That is, the diffusion tension of the solute is greater from a higher concentration to a lower than from a lower to a higher. But the diffusion tension of the solvent is greater in the direction from the lower concentration to the higher. This is also true in the case of gas and liquid solutes. When reference is made to the "concentration" of a solution, the concentration of the solute is always meant. 20 DIFFUSION AND OSMOTIC PRESSURE IV. TERMINOLOGY FOR SOLUTIONS OF DIFFERING CONCENTRATION To designate different concentrations of solutions, the most common method among physiological writers has been, until quite recently, that of percentage. An example of this method will explain its use. A solution is said to be a 5 per cent, solution of a certain solute in a certain solvent when it is composed of five parts by weight of solute to ninety-five parts by weight of solvent. But solutions of different solutes in the same solvent depend for their physical properties upon the relative number of solute particles which . they contain per unit volume. A glance at a table of atomic weights will make it clear that any method by weights which has as its basis the percentage system cannot readily be adapted to a discussion of the relative number of molecules contained in equal volumes of solutions of different solutes. Atomic weights, and therefore molecular weights, cannot readily be compared in terms of percentage. As long as physiologists persist in using this antiquated method in the preparation of their solutions, so long will they fail to arrive at any far- reaching principles concerning the chemical and physical nature of the substances used. A more scientific method is that based on the relative number of particles of solute in unit volume of solution. We cannot, of course, actually count the molecules of any substance, but from a knowledge of the relative weights of the molecules of different bodies it is easily possible to get several masses of different substances, each of which will contain approximately the same number of molecules. The weights of such masses must be to each other as the molecu- lar weights of the respective substances. For instance, 342 grams of cane sugar (mol. wt. 342) must contain the same number of molecules as 180 grams of glucose (mol. wt. 180), for the molecular weights give the relative weights of the LIQUID SOLUTIONS 21 two different molecules. Now, if these two masses are placed in equal volumes of solution, both solutes ought to show the same diffusion tension. This, indeed, is found to be true, and the same principle has been shown to be true, as far as experiment has gone, for solutions of all substances which do not conduct electricity (non-electrolytes). Solu- tions of non-electrolytes which contain the same number of molecules per unit volume have the same diffusion tension (at the same temperature) and are physically similar. Solu- tions which conduct electricity exhibit this principle only in a general way. Their departures from it and the reasons therefor will be discussed in the next chapter. The number of grams of a substance represented by its molecular weight is called a gram-molecule. Gram-molecules of all substances contain, then, the same number of molecules. If a gram-molecule of some substance be put into solution, and then this be diluted to one liter, there results a solution which can reasonably be used as a standard. Such a solution is often termed a molecular solution. Thus, a molecular solution of potassium nitrate (mol. wt. 101) is 101 grams of the salt in a liter of solution. It is as though the substance had been vaporized and the resulting gas occupied a volume of one liter. But the analytical chemist has found it convenient to use another solution as a standard. He dissolves, to form a liter of solution, as many grams of the substance in question as will react chemically with a gram -molecule of a monovalent compound. This amount of substance is termed a gram- equivalent. A gram-equivalent of sulphuric acid (H2SO4) will just neutralize a gram-equivalent of potassium hydroxide (KOH) or will just decompose a gram-equivalent of sodium chlorid (NaCl) ; but it takes two gram-equivalents of either of the last-mentioned compounds to react completely with a gram-molecule of sulphuric acid. It follows that "gram- 22 DIFFUSION AND OSMOTIC PRESSURE equivalent" and "gram-molecule" are synonymous terms in the case of monovalent compounds, and that a gram- equivalent of a bivalent compound is one half of its gram- molecule, of a trivalent compound one-third, etc. A solution made up so as to contain in one liter a single gram-equivalent of solute is termed an equivalent normal, or simply a normal, solution. Unfortunately there is a usage which terms a molecular solution normal, thus giving rise to ambiguity for all but monovalent solutes. This ambi- guity can be avoided only by the careful definition of the term "normal" by each author using it.1 For all neutral organic compounds, such as the sugars, and also for monova- lent electrolytes, a gram-equivalent is the same as a gram- molecule, and a normal solution must be a gram-molecule in a liter volume. Thus the sugar solutions described in a pre- vious paragraph are both normal solutions. No ambiguity can arise from the use of the term in reference to such com- pounds. Regarding acid salts (such as KHSO4, for example), there is a difference of opinion as to what should be denoted by gram-equivalent. Some hold (e. g., Kahlenberg) that in the salt just mentioned gram-equivalent and gram-molecule are identical. Thus, such a salt might be regarded as a monobasic acid. On the other hand, Button, Fresenius, Dandeno, and others regard a gram -equivalent of KHSO4 as one-half a gram-molecule. Thus, an equivalent solution of this salt would contain only one-half gram of hydrogen ; the salt is to be regarded as a monobasic acid, one-half of whose hydrogen has been replaced. It seems that the latter is the more truly scientific position. iFor an account of confusion (partly imagined) which has arisen from a lack of attention to such definition of these terms, see J. B. DANDENO, " The Application of Normal Solutions to Biological Problems," Bot. Gaz., Vol. XXXII (1901), pp. 229-37. Also see "Open Letters," one from Louis KAHLENBEBG, and an answer from DAN- DENO, ibid., p. 437. CHAPTER IV IONIZATION I. IONIZATION OF GASES FEOM the hypothesis of Avogadro it would be expected that, if a gram-molecule of ammonia and a gram-molecule of ammonium-chlorid vapor were put into chambers of the same size, the pressures exhibited at the same temperature by the two gases would be equal. The latter substance, how- ever, shows a far greater pressure. Now, since the kinetic theory supposes that gas pressure is due to the kinetic energy of its particles, and that the kinetic energy of any particle is dependent only upon the temperature to which it is subjected, we must either reject the theory when we come upon such a case as that just cited, or we must conclude that there are, in the mass of ammonium chlorid, a greater number of particles than in that of ammonia. Several lines of experiment and of reasoning seem to point to this as the true condition of affairs. The number of molecules is the same in both masses of gas, but in the ammonium chlorid it is supposed that many of the molecules split apart into ammonia and hydrochloric-acid ions (NH3 and HC1), and that, for producing pressure, the ions are as active as would be the same number of molecules. In this way, if all the molecules were dissociated, the pressure should be twice that required by the theory. The ammonium molecule seems not to dissociate at ordinary temperatures. Many gases exhibit the phenomenon just described ; usually ioniza- tion is not nearly complete and the pressure is simply raised above its theoretical value. As the gas becomes more con- centrated, dissociation becomes less and less complete. 23 24 DIFFUSION AND OSMOTIC PRESSURE The theory just given is called the theory of dissociation. There are other theories to account for these phenomena, but this has the widest acceptance at the present time and serves the purpose of the physiologist better than any other yet advanced. II. IONIZATION OF SOLUTES IN LIQUID SOLUTIONS It has been found that a phenomenon similar to the one just described occurs in dilute aqueous solutions of electro- lytes. These solutions uniformly give a higher diffusion tension than the one required by the theory. The explana- tion is the same as that given above ; l if ammonium chlorid, for instance, be put into aqueous solution, it has been shown that some of the molecules ionize, thus increasing the diffusion tension of the solute. The more dilute the solution, the greater is the proportion of molecules dissociated, and at infinite dilu- tion a limit would be reached at which complete dissociation would occur. This theory is named for its originator, Arrhe- nius. In very weak solutions it is found that practically all the molecules are ionized. If several electrolytes are contained in the same solution, ionization occurs in all of them, but not always to the same degree as if there were but one solute ; the presence of other molecules and ions seems to influence the amount of dissociation. This subject has not yet been sufficiently investigated to permit the formulation of a general principle. Where a solution contains several different kinds of ions, it is found that the velocity of diffu- sion of some ions is much greater than that of others. Only those substances which conduct electricity when in solution are dissociated. The whole theory of primary bat- teries and of electric conduction by liquids depends upon this principle. Enough has been said, however, to prepare the way for what is to follow. i S. AEKHENIUS, "Ueber die Dissociation der in Wasser gelOsten Stoffe," Zeit- schr. f. physik. Chem., Vol. I (1887), pp. 631-48. CHAPTER V OSMOTIC PHENOMENA I. OSMOTIC PRESSURE OF THE SOLUTE a) Non-electrolytes. — If a parchment-paper bag be filled with aqueous sugar solution and, after the opening has been sealed, the bag be submerged in water, the walls will soon be distended by an internal pressure. If the original solution is strong enough, the walls will be stretched to their limit of extensibility, and at last ruptured. If, during the distension of the bag, the water around it be tested, it will be found to be nearly or quite free from sugar ; after the bag is ruptured, however, we find the sugar diffusing rapidly to the limits of the water. Therefore parchment paper hinders greatly the diffusion of the sugar, i. e., it is only slightly permeable to dissolved sugar molecules. This fact forms the basis for an explanation of the phenomenon of distension and rupture just mentioned. In tending to dif- fuse indefinitely, the dissolved molecules (there is no disso- ciation in the case of sugar and other non-electrolytes) bom- bard the walls of any chamber in which they may be inclosed.1 The fact that they possess this property of in- definite diffusion only when within the limits of the solvent makes it necessary that such a chamber be surrounded by the pure solvent, and that the solvent permeate its walls. The pressure thus produced upon the walls of the bag is really the diffusion tension of the solute. If diffusion could take place without obstruction, this pressure would not be made apparent, but would exist none the less. The water itself exerts but little pressure upon the walls of the bag, since 1 J. H. VAN'T HOFF, " Die Rolle des osmotischen Druckes in der Analogic zwischen LOsungenund Gasen," Zeitschr. f. physik. Chem,, Vol. I (1887), pp. 481-508. 25 26 DIFFUSION AND OSMOTIC PRESSURE these are almost freely permeable to it, but a diffusion ten- sion of water exists and can be demonstrated in other ways. The internal pressure of dissolved sugar molecules forces the walls of the bag outward through the water, just as a cloth bag may be distended under water by the expansion of wire springs inside of it. Of course, in such a process of dis- tension water enters the bag from without, the bag being permeable to that substance. If in either case the bag is not strong enough to bear the pressure, it will burst when its limit of extensibility is reached. If the parchment bag is strong enough to withstand the pressure developed within, an equilibrium will be established, just as in the case of the expanding gas described in chap. ii. In this condition of equilibrium the inward pressure (due to the resilience of the walls of the bag) is just equaled by the outward pres- sure (due to the bombardment of the walls by the solute particles). Such a membrane as parchment paper, which allows the solvent to pass, but greatly retards or prevents the passage of solute particles, is said to be semi-permeable; and the pressure which such a membrane makes evident under the conditions just described is the osmotic pressure of .the solute. This is merely the diffusion tension of the solute, made evident by the opposition of the membrane. All pos- sible gradations exist between membranes which are freely permeable to solutes and those which retard them or are impermeable to them. It needs to be noted here, however, that a theoretically perfect semi-permeable membrane has not been found. The best ones which have been tested allow some passage of solute particles. Many animal mem- branes are nearly semi-permeable in certain solutions, pig's bladder being often used. A membrane of copper ferro- cyanid is almost perfectly semi-permeable in aqueous sugar solution, but it is permeable to certain salts, e. g., potassium OSMOTIC PHENOMENA 27 nitrate. The term "semi-permeable," therefore, must be used with reference to a particular solute and its solvent. Such membranes as that of copper ferrocyanid are termed "precipitation membranes;" they are formed by precipitation from two solutions which react chemically. If a solution of potassium ferrocyanid and one of copper sulphate be brought together within the walls of a porous clay cup, such a membrane (composed of copper ferrocyanid, Cu2Fe (CN)6) will be precipitated within the clay walls. The mem- brane is then supported by the clay, and the whole cup may be used for osmotic determinations. Another precipitation membrane is that formed by gelatin and tannic acid. 6) Electrolytes. — Osmotic pressure is found to be abnor- mally high in solutions of electrolytes. This is one of the facts from which the conclusion was drawn that in these solutions the diffusion tension of the solute is abnormally great, and hence that dissociation occurs. When the amount of ionization which takes place in any solution is taken into account, it is found that these solutions are only apparent exceptions to the general rule of osmotic pressure. In this case we can no longer say that the pressure is due to the bombardment of the membrane by the solute molecules, but by the solute particles, meaning thereby both molecules and ions. Solutions having the same number of solute particles per unit volume have, at the same temperature, the same osmotic pressure. As far as it has been carefully tested, this principle has been found to hold for all somewhat dilute solutions. c) Colloids. — According to their behavior when in solu- tion, substances have been classified as crystalloids and col- loids. Crystalloids produce an osmotic pressure which is practically equal quantitatively to the gas pressure which would be produced by the same number of gas particles as there are of solute particles, occupying the same volume as 28 DIFFUSION AND OSMOTIC PRESSURE the solution and possessing the same temperature. There is, however, a group of substances soluble in water, which do not produce osmotic pressure at all, or produce it in a very slight degree. These are the so-called colloids, such as gelatin, gum, silicic acid, aluminium hydroxid, etc. These substances have very large molecules which diffuse with exceeding slowness and seem to encounter great resistance in passing through water. Colloids in their relation to crys- talloids bid fair to become very important in the advance of physiological knowledge. d) Osmotic pressure in general. — Most solutes which produce osmotic pressure when in solution are either solids or liquids at ordinary temperatures, when not in solution. But a gas in solution may also produce osmotic pressure if a suitable membrane is employed. Osmotic pressure being, in its origin, perfectly comparable to gas pressure, the various principles established for gas pressure have been found to hold for osmotic pressure. The principles of Boyle, of Gay-Lussac, and of Avogadro, devel- oped for gases, have all been extended so as to include sub- stances in solution. Here is convincing evidence that a solute, as long as it is in solution, is essentially a gas occu- pying the volume of the solution. The solvent merely pro- vides conditions under which the pseudo-vaporization which we call solution can take place. Osmotic pressure is inde- pendent of the solvent and is dependent only upon the num- ber of particles of solute (i. e.^ its concentration) and upon their kinetic energy (i. e., their temperature). The nature of the solute is immaterial, the number of particles (mole- cules or ions) per unit volume being, as far as is known, the only essential factor. Where several substances that do not react chemically are in very dilute solution, the osmotic pressure of the mixture is the sum of the pressures which would be exhibited were OSMOTIC PHENOMENA 29 each of the different solutes dissolved separately to form a volume of solution equal to the original volume. As in gases, these latter pressures are termed partial pressures. This principle may be formulated again as follows : The total osmotic pressure of a dilute solution of mixed solutes is the sum of the partial osmotic pressures of the component solutes. As was seen in the last chapter, however, for more concentrated solutions this principle does not seem to hold. But more work needs to be done here before we may be positive. The principles of Boyle and Gay-Lussac would hold per- fectly true only for an ideal gas, i. e., one without any fric- tion between its particles. Such a gas does not exist, although hydrogen approaches this condition very nearly. These principles, however, hold very nearly true for all ordinary gases as long as they are not nearing the point of condensa- tion into a liquid. But, as has been stated, in the vicinity of this critical point, whether it be approached because of increase in pressure or fall in temperature, they do not hold true. In the study of osmotic pressure it is found that a similar breaking down of the same principles occurs when the solute becomes too concentrated. At high concentration the principles of gas pressure no more apply to osmotic pres- sure than they do to gas pressure itself. Just what is the action of the membrane in osmotic phe- nomena is not known. In many respects it acts like a sieve or filter, to prevent the passage of large particles, but allow smaller ones to go through unhindered. In some cases, however, the chemical nature of the membrane seems to come into play ; it seems to react chemically with the solute particles, taking them up on one side and giving them off on the other. But for a discussion of the principles and general phenomena of osmotic pressure, a knowledge of the exact method by which the membrane acts seems not to be 30 DIFFUSION AND OSMOTIC PRESSURE essential. A brief discussion of the different theories which have been proposed to account for the action of the membrane will be presented in connection with the treatment of the protoplasmic membranes of vegetable cells (see p. 80). II. DIFFUSION TENSION OF THE SOLVENT The diffusion tension of the solvent has been mentioned several times during the discussion of solutions, but it is thought well to bring together in this place the ideas con- cerning it. If the vapor tension of pure water be determined and then that of an aqueous salt or sugar solution, it will be found that the latter is invariably less than the former, and this in proportion to the concentration of the solution. Therefore it must be that particles of the solute hinder the escape of the solvent molecules. The moving particles of solute perhaps bring this about merely by moving into the path of solvent molecules which would otherwise leave the liquid. It is probable that there exists also an attraction between the particles of solute and solvent. If two solutions whose concentrations are different be brought into direct contact, as by placing a weak sugar solution over a stronger one, phenomena similar to those just discussed may be detected. Water molecules pass through the common surface in both directions. They are not vaporized, for they remain in the liquid state, but they diffuse as liquid molecules. Under such conditions the dif- fusion of the solvent is always found to be greater from the weaker to the stronger solution than in the opposite direc- tion; it will be remembered that the most rapid diffusion of the solute takes place from stronger to weaker solution. There is, therefore, a difference in the energy of diffusion (diffusion tension) of the solvent in the two solutions. This corresponds to their difference in vapor tension just de- OSMOTIC PHENOMENA 31 scribed. The diffusion tension of the solvent is greatest in the pure solvent and decreases as the concentration of the solution increases. If there were available a membrane permeable to the solute, but impermeable to the solvent, this diffusion tension of the solvent might be directly measured. It would be an osmotic pressure similar to that occasioned by the solute molecules, but of much greater magnitude and in the opposite direction. Though there is no membrane which will make this pressure evident, most of the phenomena of osmotic pressure show that it exists. No membrane can allow the solvent particles to pass absolutely without fric- tion. Thus the question arises: Why is not this pressure of the solvent made evident to a degree equal to the amount of force needed to overcome this friction? The answer is, obviously, that the pressure produced by the solvent on one side of the membrane is practically equaled by that on the other side. In solutions where the principles of osmosis hold true, the dilution of the solvent, due to the presence of the solute, is negligible. The following explanation of osmotic pressure has been given by various authors. The quotation is from Davenport : * "Upon the side containing the greater number of molecules of salt [solute] fewer water [solvent] molecules will in a given time strike the membrane than upon the other side; and since the number passing through is proportional to the number striking, relatively fewer molecules of water will consequently pass out, and so there will be a resultant flow of water to that side; and if the mass of water is confined, it will exert great pressure." This explanation is untenable for several reasons. Not nearly all solutions occupy more space than the original mass of pure solvent from which they were prepared. If to exactly a liter of water be added a 1C. B. DAVENPORT, Experimental Morphology, Vol. I (New York, 1897), p. 71. 32 DIFFUSION AND OSMOTIC PRESSURE given quantity of some solute, it cannot be told a priori whether the resulting solution will occupy the same volume as the original solvent, or a greater or less volume. Into this matter it is unnecessary to go farther than to add that osmotic pressure may be demonstrated as readily in solu- tions occupying less volume than the original solvent as in those occupying more. It is obvious that in the former case there must be a greater number of solvent particles per unit volume than in the pure solvent. Hence, if the above explanation can be retained, there should be no osmotic pressure developed in such a solution; indeed, it should appear on the side of the pure solvent. But even if it were possible that the entrance of solvent particles into the solution was due to such a difference in concentration of the solvent on opposite sides of the mem- brane, the explanation just quoted would fail. It is incon- ceivable that the osmotic membrane should be more permeable to solvent particles moving in one direction than to those moving in the other, and it thus becomes impos- sible to suppose that solvent particles which can pass the membrane in one direction "will exert great pressure" upon it in the other. Great hydrostatic pressure cannot be main- tained in a sieve, nor can osmotic pressure be maintained upon a membrane by a solvent to which it is permeable. Any difference between the solvent pressures on the two sides of an osmotic membrane must be very rapidly destroyed by diffusion of the solvent through the membrane. III. EXPERIMENTAL DEMONSTRATION OF OSMOTIC PRESSURE If a parchment-paper bag, like the one used in the illus- tration of osmotic pressure, were fixed in a firm cage, so that it could not expand except on one side, and were then filled with solution and submerged in pure solvent, bulging would occur on the free side. It would make no difference whether OSMOTIC PHENOMENA 33 this free side were serving as an osmotic membrane or not, for pressure produced anywhere in the bag must be trans- mitted equally and undiminished to all parts of the surface, in accordance with Pascal's principle of transmission of pressure in fluids. This transmission would be accomplished by the fluid as a whole, solvent and solute acting together. If, however, the portion where the bulging is supposed to take place be permeable to the solvent, the immediate pres- sure which affects the membrane must be due to the solute. Thus it comes to the same end if we consider the pressure as transmitted by the solute, acting like a gas ; for increased energy of solvent particles will be transmitted to the solute particles with which they come in contact. A very common mode of demonstrating the existence of osmotic pressure is the following : A piece of animal bladder or parchment paper is tied tightly over the expanded end of a thistle tube, and the bulb is filled with molasses, or a strong aqueous sugar or salt solution. Then the tube is fastened upright, the bulb being immersed in water so that the liquids within and without have a common level. After a time it is observed that the solution has risen in the tube, often to the height of a meter or more if the tube is sufficiently long. The diffusion tension of the solute particles within the bulb is of course operative in every direction, but osmotic pressure is developed and made apparent only within the membrane. This pressure is transmitted through the liquid to all parts of the surface of the solution. But the only part of this surface which is free to move, after the limit of extensibility of the membrane is reached, is the free surface in the stem of the tube. We have seen that the free surface of a liquid is bounded by a peculiar layer or film. Upon this film the transmitted osmotic pressure is effective, just as though the film were a piston closely fitting within the bore of the tube. In this case, since the surface layer is nearly impermeable to 34 DIFFUSION AND OSMOTIC PRESSURE liquid solvent, the pressure of solvent particles may be imme- diately effective. The surface film is lifted by the pressure exerted from below, as a piston in such a position might be lifted by wire springs situated within the bulb; and, in rising, be the change in level ever so slight, it increases the volume of the solution in the tube, thus decreasing the dif- fusion tension of the solvent within this solution, and also overcoming to some degree the atmospheric pressure on the free surface of the solution; and water enters through the permeable membrane below. The entrance of the water is due mainly to the diffusion tension of the solvent and in part to hydrostatic pressure. In the latter sense the push- ing up of the surface film acts like the raising of a piston in a pump. With a closed water manometer, or an open one of mercury, a pressure far surpassing that of an atmosphere may be obtained where the membrane used is sufficiently strong. Of course, in such cases hydrostatic pressure as a cause for the ascent of the column is to be ruled entirely out of consideration. Other methods of demonstrating osmotic pressure are in use, but the explanation just given may be applied, mutatis mutandis, to any of them. CHAPTER VI MEASUREMENT AND CALCULATION OF OSMOTIC PRESSURE I. MEASUREMENT OF OSMOTIC PRESSURE a) Direct method. — Tha direct method of measurement of osmotic pressure is very difficult of operation, and deter- minations thus made are exceedingly tedious processes. Nor is this method susceptible of sufficient accuracy to recommend it to physiologists. But since it is the classical method used by Pfeffer1 in his original investigation of the subject, and since it has been used since that time by physical chemists in establishing the principles by which indirect methods become available, it will be described here at some length. A membrane of copper ferrocyanid is precipitated within the walls of a cup or bulb of porous clay (a filter bulb serves admirably) by filling the bulb with a solution of po- tassium ferrocyanid and surrounding it with one of copper sulphate. The bulb should first be thoroughly cleaned and freed from air by boiling for some time in water. When the membrane is well formed (which occurs after fifteen to forty hours), the cup is filled with the solution to be tested, closed with a rubber stopper bearing a mercury manometer, and immersed in water. The osmotic pressure rises for a number of hours, being indicated by the rise in the mercury column, and at last, when the membrane has been ruptured somewhere, begins to descend again. The maximum read- ing of the mercury column is taken as the osmotic pressure of the solution. The difficulty of the method lies in getting i W. PFEFFEK, Osmotische Untersuchungen^ Leipzig, 1877. 36 DIFFUSION AND OSMOTIC PRESSURE a perfect semi- permeable membrane which will withstand the high pressures developed. A number of determinations for the same solution are necessary in order to eliminate erratic cases. The direct method of measurement has just been brought again into prominence by the work of Morse and Horn.1 These authors have succeeded in forming much more perfect membranes in porous clay cups than have ever been pro- duced before. Air is first swept out of the pores of the cup by an "endosmotic" current. The cup is filled with a weak solution of K2SO4 and immersed in a vessel of the same solution until the outer level is near the margin of the cup. Then a current from a dynamo is passed between a cylindri- cal copper electrode surrounding the cup and a platinum electrode within it. As the liquid rises in the cup, it is removed, and in a short time the air is all removed from the porous clay. Then the cup is filled with K4Fe(CN)6 and immersed as before, but now in a solution of CuSO4. The current is passed again, and thus the Fe(CN)6 ions are driven into the clay from one side, while the Cu ions are forced in from the other. The resistance of the cup gradu- ally rises as the membrane is formed, being from fifteen hundred to three thousand ohms at the time when the mem- brane is considered as complete.2 b) Indirect methods. — Owing to the difficulties encoun- tered in the use of the direct method just described, an indirect method is usually resorted to. These indirect methods depend upon the general principles, that depression 1 H. N. MORSE AND D. W. HORN, "The Preparation of Osmotic Membranes by Electrolysis," Am. Chem. Jour., Vol. XXVI (1901), pp. 80-86. 2 The most recent investigations into the nature of precipitation membranes are as follows : G. TAMMANN, " Ueber die Permeabilitat der Niederschlags-Membranen," Zeitschr. f. physik. Chem., Vol. X (1892), pp. 255-64; P. WALDEN, "Ueber Diffu- sionserscheinungen an Niederschlags-Membranen," ibid. (1892), pp. 699-732; J. H. MEERBURG, u Zur Abhandlung Tammanns : Ueber die Permeabilitat der Nieder- schlags-Membranen," ibid., Vol. XI (1893), pp. 446-8. MEASUREMENT AND CALCULATION 37 of the freezing-point, elevation of the boiling-point, decrease of the vapor tension of solutions, and osmotic pressure are all related phenomena, and may be obtained one from the other for any given solution.1 Three of the most satis- factory methods for determining osmotic pressure in this way will be briefly described here : 1. The freezing-point method : The freezing-point of a solution is always lower than that of the pure solvent. This depression of the freezing-point is proportional to the num- ber of solute particles present, and therefore to the osmotic pressure. The depression of the freezing-point can best be deter- mined by means of Beckmann's apparatus,2 which may be found described in any of the texts on physical chemistry. A determination of the freezing-point is first made for distilled water ; this is followed by a determination for the solution to be tested, care being taken not to disturb the adjustment of the thermometer between the determinations. The difference between the two observations will be the required depression, which may be denoted by A^. The rela- tion between this quantity and the osmotic pressure is ex- pressed, for aqueous solutions, by the following equation : Pf= 9173.2 A, ,3 wherein Pf is the osmotic pressure at the freezing-point of the solution, measured in millimeters of mercury. The osmotic pressure at any desired temperature other than the freezing-point, say T in the absolute scale, may be obtained by applying the principle of Gay-Lussac, which 1 J. H. VAN'T HOFF, " Die Rolle des osmotischen Druckes in der Analogic zwischen LOsungen und Gasen," Zeitschr.f. physik. Chem., Vol. I (1887), pp. 481-508. 2E. BECKMANN, "Ueber die Methode der Molekulargewichtsbestimmung durch Gefrierpunktserniedrigung," ibid., Vol. II (1888), pp. 638-45. 3 NERNST-PALMER, Theoretical Chemistry (London, 1895), p. 132. The pressure here is reduced from atmospheres to millimeters of mercury. 38 DIFFUSION AND OSMOTIC PRESSURE holds for osmotic pressures of dilute solutions. This opera- tion is expressed in the following: PTTf=PfT , in which PT is the osmotic pressure, in millimeters of mer- cury, at required temperature T (absolute), and Tf is the absolute freezing-point of the solution. From the equation we get: T P* = P>T, • In the case of weak aqueous solutions, the freezing-point of the solution may be considered, for this calculation, as prac- tically the same as that of the solvent. Thus Tf =273° (the freezing-point of pure water), and T becomes 273 + /, where t is the desired temperature in the Centigrade scale. Now the equation given above becomes: P = = pf(l + ^L t^ = Pf (i + 0.00367 1) . This is sufficiently accurate for dilute aqueous solutions. The freezing-point method is the simplest and most satis- factory method for general use. 2. The boiling-point method: The boiling-point of a solu- tion is always higher than that of the pure solvent, and its elevation is proportional to the osmotic pressure at that tem- perature. The relation between the two quantities for aqueous solutions is expressed as follows: wherein Pb is the osmotic pressure in millimeters of mercury at the boiling-point of the solution, and A6 is the elevation of the boiling-point. The determination of the boiling- point of the solution and of distilled water is best made i NEBNST-PALMEB, Theoretical Chemistry (London, 1895), p. 129. The pressure is again reduced to millimeters of mercury. MEASUREMENT AND CALCULATION 39 by Beckmann's improved apparatus for this purpose,1 a de- scription of which will be found along with that for the freezing-point determinations. The correction for temperature may be made, as in the last case, by the application of the principle of Gay-Lussac directly, or by interpolation between Pb and Pf, the latter having been determined by the previous method. The expression for the Gay-Lussac principle is of course the same, mutatis mutandis, as that given above: T P J T — -MJ^T 5 in which PT is again the pressure in millimeters at the de- sired temperature T, in the absolute scale, and Tb is the absolute boiling-point of the solution. For this calculation the boiling-point of a weak aqueous solution may be con- sidered the same as that of pure water. Thus Tb = 373°, the boiling-point of water, and T = 273 -f- £, where t is the desired temperature in the Centigrade scale. Making these changes in the above equation, 273 + * 6 373 The method of interpolation is expressed by the follow- ing equation: where Pt is the pressure at the desired temperature, t (Centi- grade), and the other symbols are the same as above. 3. Method by observed vapor tension: As has already been stated, the vapor tension of the solvent is decreased by the presence of a solute. It is found that, for dilute solu- tions, this decrease in vapor tension is proportional to the i E. BECKMANN, " Zur Praxis der Bestimmung von Molekulargewichten nach der Siedemethode," Zeitschr.f. physik. Chem., Vol. VIII (1891), pp. 223-8. 40 DIFFUSION AND OSMOTIC PRESSURE osmotic pressure. This relation is expressed by the follow- ing equation: TT-TT' 0.0819 T X 1000 S X 760 l 77 ~1F~ In this P is the osmotic pressure in millimeters at the abso- lute temperature T, TT and TT ' are the vapor tensions observed at that temperature of the solvent and solutions respectively, s is the specific gravity of the solution, and M is the molecu- lar weight of the pure solvent. In the case of dilute aqueous solutions, s may be put equal to unity (the specific gravity of the pure solvent instead of that of the solution), and M is 18 (the molecular weight of water). Making these substitutions in the above equation, we have: TT-TT' 0.0819 Tx 1000 X 760 7T 18 or The determination of the vapor tensions is best made by means of the method devised by Ostwald and Walker.2 Two Liebig potash bulbs, one filled with the solution to be tested and the other with the pure solvent (the latter weighed) , are joined in series and then attached to a weighed U-tube of pumice moistened with sulphuric acid. A slow current of air is passed, for six to twelve hours, through the series. The air first becomes saturated at the tension of the solution, and then, passing through the second bulb, be- comes again saturated at the vapor tension of the pure solvent. A final weighing of the second bulb and of the 1 NEKNST-PALMEK, Theoretical Chemistry (London, 1895), p. 126; also J. H. VAN'T HOFF, " Die Rolle des osmotischen Druckes in der Analogie zwischen LOsun- gen und Gasen," Zeitschr.f. physik. Chem., Vol. I (1887), pp. 481-508. 2 J. WALKER, " Ueber eine Methode der Bestimmung der Dampfspannung bei niederen Temperaturen," Zeitschr. f. physik. Chem., Vol. II (1888), pp. 602-5; also OSTWALD-WALKEB, Manual of Physico-Chemical Measurements (London, 1894), p. 188. MEASUREMENT AND CALCULATION 41 sulphuric-acid tube gives the required data. The amount of vapor removed from the two bulbs respectively is proportional to the vapor tensions of their contents. Thus if w denote the loss in weight in the second bulb and w' the gain in weight of the sulphuric acid, w TT — TT' ^7" ~H~ ' Therefore the equation given above may be written: w This method is difficult of operation and not very satis- factory. The whole apparatus must be surrounded by a jacket to keep all the parts at the same temperature ; it is not necessary that the temperature be absolutely constant, however. The only advantage in this method over those previously described is that by this means the osmotic pressure can be determined for the temperature at which the solution is used, thus avoiding the correction for tem- perature. II. CALCULATION OF OSMOTIC PRESSURE a) When the pressure is produced by a non-electrolyte. — All solutions of non-electrolytes which contain the same number of molecules per unit volume of solution give the same osmotic pressure. From measurements made by Pf efifer we know that the osmotic pressure of a solution of sugar containing a gram-molecule per liter is the same as the gas pressure of a gram-molecule of gas occupying a liter volume. This pressure is 22.3 atmospheres, or 16,948 mm. of mercury, at 0° C., or 273° absolute. Thus, if we know the molecular weight of the solute and the number of grams per liter of solution, the calculation, on the principle that pressure varies as concentration, is simple enough. The correction for tem- perature is carried out by the principle of Gay-Lussac. 42 DIFFUSION AND OSMOTIC PRESSURE For mixed solutions of non-electrolytes the total osmotic pressure is the sum of the partial pressures due to the several solutes respectively. b) When the pressure is produced by an electrolyte. — On account of the phenomena of ionization or dissociation, the calculation of the osmotic pressure of a solution of an electrolyte becomes somewhat complicated. The amount of ionization must be known 'in order to get the relative number of particles per unit volume. For instance, a gram- molecule of NaCl, in aqueous solution, occupying a liter volume, contains more particles than the same volume of a normal solution of sugar ; some of the molecules have sepa- rated into Na and Cl ions. The amount of ionization in any simple solution of an electrolyte can be determined by means of the method of electrolytic conductivity devised by Kohlrausch.1 The con- ductivity is proportional to the number of free ions, and hence, knowing the conductivity both at the given concen- tration and at a concentration where ionization is complete, we can calculate the amount of ionization. The conductivity of many solutions has been determined by different authors in different units. Of course, all are reducible to C. G. S. units or to the conductivity of mercury, but it is immaterial for the present purpose what units are used, so long as the same ones are used throughout the same calculation. As the solution becomes more and more dilute, the conductivity approaches a limit. This limit is the conductivity at infinite dilution, where ionization is complete; it is usually denoted by Xw. Allow X to denote the conductivity at the given concentration. Then — = a , the fraction of the whole Aoo IF. KOHLRAUSCH, Leitfaden der praktischen Physik, 7th ed. (Leipzig, 1892), p. 301 ; also F. KOHLRAUSCH UNO L. HOLBORN, Das LeitvermOgen der Elektrolyte, Leipzig, 1898; W. OSTWALD, "Ueber Apparate zur Bestimmung der Electrischen Leitfahigkeit von Electrolyten, Zeitschr. f. physik. Chem., Vol. II (1888), pp. 561-67; OSTWALD-WALKER, Manual of Physico-Chemical Measurements, London, 1894. MEASUREMENT AND CALCULATION 43 number of molecules which are dissociated. Thus, if one out of every ten molecules were dissociated, a would equal T^. Now, if each molecule forms k ions, and if i denote the ratio of the actual osmotic pressure to that which would be obtained in the same concentration of a non-electrolyte solution, i may be found from the following: t = l + (fc-l)a . And if P denote the osmotic pressure developed in a solution of a non-electrolyte, of the same concentration as that whose pressure is to be found, P' being the required pressure, then P' = Pi . The conductivities of a great many solutions are to be obtained from published tables.1 It is not necessary to give the methods for determining these conductivities here. They are thoroughly and completely discussed by Kohl- rausch and Holborn. If the osmotic pressure is all that is required, and data for the conductivity of the given solute cannot be found in the published tables, then it is more expe- dient to determine the pressure by means of one of the indirect methods previously described than to determine the conduc- tivity. If the proper concentration is not given in the tables, the conductivity for it is found by interpolation between the conductivities for the two concentrations nearest to it. If the table is rather extensive for the solution in question, so that conductivities for very low concentrations are given, it is usually safe to take the highest conductivity as \M. If the table is not so complete, a limit for the con- ductivity has to be approximated from the trend of the given data. In using the published tables, it is very important that one bear in mind the difference between molecular and i W. C. D. WHETHAM, Solution and Electrolysis (Cambridge, 1895), pp. 218 ff.; also F. KOHLBAUSCH UND L. HOLBOKN, Das Leitvermogen der Elekrolyte, Leipzig, 44 DIFFUSION AND OSMOTIC PRESSURE equivalent solutions. Most of the tables consider as a standard solutions containing a gram-equivalent per liter. These are easily transformed into gram-molecular solutions by dividing the given concentration by the number repre- senting the valency of the compound. Thus one-tenth gram-equivalent per liter of Na2SO4 is identical with one- twentieth gram-molecular solution of the same salt. For very weak solutions of mixed electrolytes the above method may be resorted to. But for solutions of mixed electrolytes and non-electrolytes, and for strong solutions of electrolytes, no method of calculation has yet been discovered. The only practical way open in such a case is to resort to the methods of freezing- and boiling-points. It is often best to make use of both these methods, and to interpolate between them for the normal temperature, inasmuch as ionization often increases rapidly at higher temperatures. Of course, where chemical reaction occurs between the different solutes, the osmotic pressure of the solution will not be constant until chemical equilibrium has been attained. PART II . PHYSIOLOGICAL CONSIDERATIONS INTRODUCTION So IMPORTANT a part do diffusion and osmotic pressure seem to play in the vital processes of plants, that it is well- nigh impossible to consider any phase of vegetable physiology without some reference to these subjects. It is obviously not to the point, however, to attempt here a discussion of every phenomenon in plant life into which they enter. Rather will attention be directed to certain groups of phenomena wherein diffusion and osmotic pressure seem to be fundamental factors. Thus, it is hoped, may be formed a general conception of the trend which modern study is taking along these lines. Of the four following chapters, the first three have to do with osmotic pressure as an internal factor in the life of the plant; in them are considered the most important effects of the development of diffusion tensions within the plant body. In the last chapter are brought together the responses of the organism to variations in the osmotic pressure of the sur- rounding medium. Such division of the subject is merely expedient; it is purely artificial, for the organism and its surrounding medium are physically almost as truly continu- ous as are a mass of ice and the water in which it floats. Also — a fact which is often apparently lost sight of — every portion of the plant body is a portion of the environment of every other portion. This is of fundamental importance, especially in the physiology of multicellular forms. How- ever, the plant body is a fairly definite thing, and in the present state of our knowledge the above classification of environmental factors is perhaps as good as any other. In the following pages authors are cited for the most 47 48 DIFFUSION AND OSMOTIC PRESSURE important pieces of research, mainly for the more recent ones. References are not given for material which may be regarded as a matter of common knowledge. To those who wish full citations for the period up to the time of its publi- cation, Ewart's admirable translation of Pfeffer's Physiology of Plants will be found of great service. CHAPTER I TURGIDITY I. PROTOPLASM AND ITS LIMITING MEMBRANES ANYTHING resembling an exact knowledge of the nature of protoplasm is very remote from us as yet, but we may be fairly certain of this, at least, that, whatever else it may be, the vital substance is a mixture of many soluble colloids dissolved in, or impregnated with, an aqueous solution of many different crystalloids. Colloids are very inactive as far as diffusion and osmotic pressure are concerned. Thus, if an internal diffusion tension is developed within a mass of protoplasm, it must be mainly due to the crystalloids dis- solved in the contained water. On this account it must come about that a mass of colloid substance inclosing within its body an osmotic solution, and surrounded by another osmotic solution, will act somewhat as though the former solution were surrounded by a semi-permeable membrane. Because of their slow rate of diffusion, colloid particles must in a measure block the way for the diffusion of crystalloid particles. Hence, if the more concentrated osmotic solution be within the colloid mass, there will be developed a slight osmotic pressure within the mass, which will hasten its normal process of swelling by imbibition. With no truly semi-permeable membrane about it, no state of equilibrium can be attained between a colloid body and the surrounding medium, until, by the slow outward diffusion of the crystal- loid particles and by the entrance of water, there comes about a uniform concentration both within and without. In the author's experiments with gelatin plate cultures of Stigeoclonium the following observations were made, which 49 50 DIFFUSION AND OSMOTIC PRESSURE appear to have a bearing in this connection: A somewhat concentrated solution of mineral salts was thickened by the addition of enough gelatin to make a firm mass at ordinary temperatures. On the surface of this mass were placed single drops of a dilute solution having the same chemical nature as the one contained within the gelatin plate, and the whole was kept in a moist chamber. After four or five hours it was always noted that the drops of liquid had disappeared ; they had been absorbed into the colloid mass. If, however, the more dilute solution were contained within the gelatin plate, and drops of a concentrated solution were placed upon its surface, it took very much longer for total absorption to occur. For the first few hours there was usually even an observable increase in the size of the liquid drops. Eventually absorp- tion occurred, but it was often at the end of a period of more than twenty-four hours. Of course, if there had been a semi-permeable membrane between the drops and the gel- atin, absorption would not have taken place. The gelatin mass is not semi-permeable, but seems merely to retard the process of diffusion of crystalloid solutes. If a mass of such gelatin, containing a strong osmotic solution and surrounded by a semi-permeable membrane, be placed in water or a weaker solution, this membrane will be stretched by the internal pressure practically as though no colloid were present, and a state of equilibrium will be reached only when the resilience of the membrane equals the osmotic pressure within. This has been demonstrated experimentally by Traube and Pfeffer.1 In such a case the osmotic pressure of the colloid is of such an order as to be negligible. Now, any mass of protoplasm is very much the same sort of a colloid mass as the gelatin just described. Its outer 1M. TRAUBE, "Experimente zur Theorie der Zellenbildung und Endosmose," Arch. f. Anat. u. PhysioL, Physiol. Abth., Jahrg. 1867, pp. 87-165. Also PFEFFEB- EWABT, Physiology of Plants, Cambridge, 1900, p. 106. TUKGIDITY 51 layer is so transformed (perhaps in many instances by mere contact with the external solution and with surrounding objects) that it is almost perfectly impermeable to many solutes, but remains permeable to water. The protoplast of every normal vegetable cell is thus surrounded by a more or less perfectly semi-permeable layer, the ectoplast. If the ectoplast is ruptured in any way, it is soon re-formed, unless disorganization of the protoplast ensues.1 In such cases (e. g., in Myxomycete plasmodia, etc.), where unmodified protoplasm is brought into contact with surrounding medium, it is perhaps partially on account of its colloidal nature that the contained crystalloids are not immediately lost by dif- fusion, instead of being retained, as they are, until a new surface layer can be formed. The internal osmotic pressure, which results when the inclosed solution is more concentrated than the external one, tends to stretch the surface layer and enlarge the protoplast. Against this pressure is brought to bear whatever cohesive and resilient force the ectoplast may possess; but this, from the semi-fluid nature of protoplasm itself, must be of a low order. In naked cells this fact prevents the internal pressure from ever becoming very great; in such cases rupture and destruction of the protoplasm would inevitably result. But if the protoplasm is surrounded by a cellulose membrane, as in the case of the majority of plant cells, this condition is entirely altered; the swelling of the protoplas- mic mass is checked at the limit of extensibility of the inclosing cellulose layer. Pressure upon the ectoplast is transmitted immediately to the cell wall, and the latter is stretched according to its extensibility and to the pressure applied. In the condition of strain resulting from the interaction of the force of osmotic pressure (the diffusion 1 W. PFEFFER, "Zur Kenntniss der Plasmahaut u. d. Vacuolen," etc., Abhandl. d. k. sachs. Ges. d. Wiss. zu Leipzig, math.-physik. Klasse, Vol. XVI (1890), pp. 187-344. 52 DIFFUSION AND OSMOTIC PRESSURE tension of the solute particles) on the one hand, and that of resilience of the cellulose membrane on the other, a rigidity and firmness of the cell as a whole is brought about, just as a football or bicycle tire becomes rigid and firm upon being inflated with gas. This rigidity is termed turgescence, or turgidity. The term "turgor" has also been applied to this condition, but it is better to reserve this word to express the osmotic pressure of the internal fluid.1 In order that the protoplast may retain its osmotic properties, the cellu- lose wall must be permeated with water. This is absolutely essential for the development of turgidity, since osmotic pressure is not active beyond the limits of the solvent. It is, indeed, true that the cellulose envelope of every active cell is saturated with water. Thus far only those cells have been considered which are completely filled with protoplasm. This is the condition in young cells, but mature cells are not usually so filled; as growth progresses, vacuoles of a watery fluid appear in the protoplasm. These increase in size and fuse together, until at length there is a single large vacuole within the protoplasmic mass. The typical cell of plant tissues consists of a cellulose wall lined internally by a layer of protoplasm, which incloses a mass of aqueous solution, the cell sap, containing sugars and various other solutes. The lining layer of protoplasm is bounded externally, where it comes in contact with the cell wall, by the ectoplast. Internally, toward the vacuole, it is bounded by a similar membrane, the tonoplast. The cellulose wall is readily permeable to water and solutes, but the protoplasmic lining, with its two somewhat differentiated limiting layers, normally acts like a semi-permeable mem- brane, allowing water to pass quite freely, but hindering, and often seeming absolutely to prevent, the passage of solutes. iCf. E. B. COPELAND, "The Mechanism of Stromata," Ann. Bot., Vol. XVI (1902), p. 330; IDEM, "The Rise of the Transpiration Stream," Bot. Gaz., Vol. XXXIV (1902), p. 173. TUBGIDITY 53 It is in these vacuolated cells that turgidity is developed to its greatest extent. It may be that continued concentra- tion of the solution within the protoplasm itself may soon reach a limit beyond which it cannot go without affecting those energy transformations which we term vital activity or life. An alteration in the activities of the protoplasm thus produced may result in a change in its permeability in one way or another. And changes of this sort accompanied by changes in the chemical activity within the protoplast may account for the formation of the vacuole and the secretion of osmotically active materials into it. It was shown by Loeb1 that changes in the concentration of different ions in the protoplasm of animal muscle bring about marked changes in its power of absorbing water. At any rate, however the vacuole may arise, the turgidity of the normal mature plant cell is mainly due to the osmotic pressure of the cell sap and to the semi-permeability of the surrounding protoplasmic layer. The part played in the development of turgidity by the tonoplast and ectoplast and by the unmodified protoplasm itself, has not been worked out. Indeed, the semi-permeability of this layer can perhaps be attained only through the co-operation of the three some- what distinct layers which make up the lining of the cellu- lose wall. Although De Vries2 was able to separate the tonoplast from the remainder of the protoplasmic mass, yet it soon lost its peculiar properties when the surrounding protoplasm was killed. Pf effer 3 has shown that the tonoplast and ectoplast are equivalent and are probably formed in the 1 J. LOEB, " On Ion Proteid Compounds and Their Role in the Mechanics of Life Phenomena"; I, "The Poisonous Character of a Pure NaCl Solution," Am. Jour. PhysioL, Vol. Ill (1900), pp. 327-38. 2 HUGO DE VRIES, " Plasmolytische Studien fiber die Wand der Vacuolen,'1 Jahrb.f. wiss. Bot., Vol. XVI (1885), pp. 465-598. The tonoplast is not a special cell organ, as De Vries was led to suppose. 3 W. PFEFFER, " Zur Kenntniss d. Plasmahaut u. d. Vacuolen," etc., Abhandl. d. k. sacks. Ges. d. Wiss. zu Leipzig, math.-physik. Klasse, Vol. XVI (1890), pp. 187-344. 54 DIFFUSION AND OSMOTIC PRESSURE same manner. In consideration of such facts as these much stability cannot be predicated of these membranes, and thus, in a discussion of the osmotic properties of the cell, it will probably be safer to regard the intra-vacuolar pressure as arising from the semi-permeability of the lining layer of protoplasm as a whole. In a vacuolated cell the osmotic pressure sometimes becomes so great as to burst the cellulose membrane. This is notably so in the bursting of the asci in certain ascomy- cetous fungi and in the explosion of the hypo-sporangial region in Pilobolus. Many plant cells may be made to burst in this way by immersing them in a very weak solution or in distilled water. For example, Lidforss1 found that the pollen grains of many plants (notably the Liliaceae, as Funkia, Asphodelus, Anthericum, etc.) exploded in this way when put into water. Noll2 has shown that when certain marine Siphoneae are placed in pure water their filaments are apt to burst. Also Curtis3 noted that when the common molds were placed in water after having been accustomed to a concentrated solution, the hyphal tips often burst in the same manner. Similar observations were made among ani- mals by Gogorza,4 who records the bursting of blood cor- puscles in certain marine forms when they were killed by being placed in fresh water. II. PLASMOLYSIS When a plant cell is surrounded by a solution of greater concentration than that contained within its vacuole, the phenomenon of plasmolysis occurs. The greater osmotic 1 B. LIDFORSS, " Zur Biologie des Pollens," Jahrb. f. wiss. Bot., Vol. XXIX (1896), pp. 1-138. 2 F. NOLL, "Beitrag 7.ur Kenntniss der physikalischen Vorgange welche den Reizkrttmungen zu Grunde liegen," Arb. d. hot. Inst. in Wiirzburg, Vol. Ill (1888), p,496. 3C. CURTIS, "Turgidity inMycelia," Bull. Torr. Bot. Club, Vol. XXVII (1900), pp. 1-13. 4 GOGORZA Y GONZALEZ, "Influencia del aqua dulce en los animales marinos," Annalesde la soc. espagn. hist, nat., Vol. XX (1891), pp. 220-71. TURGIDITY 55 pressure of the solutes outside, together with the slight resilience of the protoplasmic layer, cause a contraction of the protoplasm resulting in its separation from the inclosing cellulose wall. If the process of plasmolysis is complete the vacuole may disappear, practically all the water passing out. In such cases the protoplasm often takes on the form of a solid sphere, which lies near the middle of the cell or at one side. Plasmolysis comes about within a very few minutes after the cell has been placed in the plasmolyzing solution. It is partly from the latter fact that the cellulose wall is known to be permeable to solutes as well as to water. If it were not so, either plasmolysis would not occur, or the cellulose membrane would follow the protoplasm in its withdrawal toward the center of the cell. The cellulose wall does, indeed, contract to a certain measurable extent, but this is due entirely to its elasticity ; it simply returns to its normal state of equilibrium when the internal pressure of the turgid protoplasmic sac is removed. This fact of plasmolysis has long been known,1 but the true interpretation of it was due to De Vries 2 and Pf effer.3 After the relation which exists between plasmolysis, turgidity, and osmotic pressure was once established, it was De Vries* who pointed out that in the former of these phenomena we possess a means of measuring the amount of osmotic pressure in any given cell. His method has been used very largely in such measurements. It may be described as follows: If a piece of plant tissue be placed in a concentrated solution 1 For a historical treatment of this subject see W. PFEFFER, " Zur Kenntniss der Plasmahaut u. d. Vacuolen," etc., Abhandl. d. k. sachs. Ges. d. Wiss. zu Leipzig, math.-physik. Klasse, Vol. XVI (1890), p. 316. 2 H. DE VRIES, Untersuchungen iiber die mechanischen Ursachen der Zellstreck- ung, Leipzig, 1877. 3 W. PFEFFER, Osmotische Untersuchungen, Leipzig, 1877, pp. 121 ff. * H. DE VRIES, " Eine Methode zur Analyse der Turgorkraft," Jahrb.f. wiss. Bot., Vol. XIV (1884), pp. 427-601. IDEM, "Osmotische Versuche mit lebenden Membranen," Zeitschr.f.physik. Chem., Vol. II (1888), pp. 415-32; IDEM, "Isotonische Koeffizienten einiger Salze," ibid., Vol. Ill (1889), pp. 103-12. 56 DIFFUSION AND OSMOTIC PRESSURE of potassium nitrate, plasmolysis will occur. If tissues with colored cell sap, such as portions of the lower epidermis of the leaves of Tradescantia, are used, contraction of the vacuole may be seen very readily under the microscope. The coloring matter of the sap fails to pass the protoplasmic layer, and thus plasmolysis is accompanied by a deepening of the color of the sap. If the experiment be repeated on fresh bits of tissue, continually weaker and weaker solutions of potassium nitrate being used, a concentration of the latter will at length be reached, such that no plasmolysis will occur. But plasmolysis indicates that the external solution is more concentrated than that within the vacuole, and its failure to appear indicates that the cell sap is more concen- trated than the external solution. Therefore, it may be considered that the maximum concentration of potassium nitrate which does not cause plasmolysis is isosmotic (i.e., has the same osmotic pressure) with the cell sap. If we can choose a plasmolyzing substance to which the protoplasmic membrane is very nearly or quite impermeable1 (see the fol- lowing section), this will give a very exact method for measuring turgor pressure. In this way De Vries was able to show that, in general, the concentration causing plas- molysis was always the same, no matter what substance was used to produce it. There were some exceptions, however, glycerin being the most notable of those used by him. He found, too, that certain electrolytes gave extraordinarily high osmotic pressures. The last is now known to be due to ionization. The " isotonic coefficients " given by this author express approximately the amount of ionization for the con- centrations which he used. The results are exceedingly valuable, for they have led to great advance, not only in physiology, but also in physical chemistry ; but since these i DE VKIES has given some very pointed directions for the critical use of his method in "Zur plasmolytischen Methodik," Bot. Zeitg., Vol. XLII (1884), pp. TURGIDITY 57 coefficients hold true only within certain limits, and since other more accurate methods are now available for deter- mining the amount of ionization, a discussion of them is here omitted.1 On the animal side the method of plasmolysis has been used by Hamburger and others 2 for determining the osmotic pressures of the fluid contained in blood corpuscles. Another method for comparing the osmotic pressure of the fluid contained in red blood corpuscles with that of the surrounding fluid was devised by Koppe,3 and further used by Lob* and Hedin.5 The total volume of all the cor- puscles of a given amount of blood was first determined by separating them from the plasma by means of the centrifuge. Then a known amount of blood was added to a given volume of salt solution of known concentration, and the mixture was shaken thoroughly. The corpuscles were then separated from the solution on the centrifuge, and their total volume carefully measured. If the resulting volume was less than the normal for the given amount of blood, the conclusion 1 It will be well for physiology when the practical use of these coefficients dies out entirely. 2 H. J. HAMBURGER, " Ueber den Einfluss chemischer Verbindungen auf Blut- kOrperchen in Zusammenhang mit ihren Molekulargewichten," Arch. /. Anat. u. Physiol., Physiol. Abth., Jahrg. 1886, pp. 476-87; IDEM, " Ueber die durch Salz- und RohrzuckerlOsungen bewirkten Veranderungen der BlutkOrperchen," ibid., Jahrg. 1887, pp. 31-47; IDEM, "Die Permeabilitat der rothen BlutkOrperchen in Zusammen- hang mit den isotonischen Coefficienten," Zeitschr. f. Biol.,Vol. XXVI (1889), pp. 414-33; G. GETNS, "Ueber d. Einfluss gelOster Stoffe auf d. rothen Blutzellen, in Ver- bindung mit d. Erscheinungen der Osmose u. Diffusion," Pflilgers Arch. f. d. ges. Physiol., Vol. LXIII (1896), pp. 86-119; H. KOPPE, "Physiologische KochsalzlOsung— Isotonie— osmotischer Druck," ibid., Vol. LXV (1897), pp. 492-502. 3 H. KOPPE, " Eine neue Methode zur Bestimmung isotonischer Konzentrati- onen," Zeitschr. f. physik. Chem., Vol. XVI (1895), pp. 261-88. * W. LOB, "Ueber Molekulargewichtsbestimmung von in Wasser lOslichen Sub- stanzen mittels der rothen BlutkOrperchen," ibid., Vol. XIV (1894), pp. 424-32. 5S. G. HEDIN, "Ueber d. Brauchbarkeit der Centrifugalkraft far quantitative Blutuntersuchungen," Pflilgers Arch. f. d. ges. Physiol., Vol. LX (1895), pp. 360- 404; IDEM, "Ueber d. Bestimmung isotonischer Konzentrationen durch Zentrifugieren von Blutmischungen," Zeitschr. f. physik. Chem., Vol. XVII (1895), pp. 164-70; IDEM, u Einige Bemerkungen KOppes Abhandlung: Ueber eine neue Methode, etc.," ibid., Vol. XXI (1896), pp. 272-6. 58 DIFFUSION AND OSMOTIC PEESSURE was drawn that the corpuscles had lost water, and hence that the surrounding solution was of higher osmotic pressure than the internal one. Several slight modifications of the method were used, and many different solutions were com- pared, the results being quite uniform with those obtained by direct observation of the cells by De Vries and Hamburger. Still another manner of comparing osmotic pressures of various solutions by means of plasmolytic phenomena in liv- ing cells is that used by Wladimiroff,1 who brought motile bacteria into requisition for the purpose. He found that these organisms cease to be motile when the osmotic pressure of the surrounding fluid attained a certain magnitude. Using as a criterion the degree of concentration at which motion ceased, he compared the osmotic pressures of a num- ber of solutions. His results are, in general, uniform with those of the other authors just mentioned. The loss of motion was due to extraction of water in a manner exactly analogous to plasmolysis. In making turgor determinations by the plasmolytic method the results may be given in various ways. The usual method has been to give them in terms of a per cent, solution of potassium nitrate, sometimes of sodium chlorid, sometimes of sugar, etc. But with this method, whenever it is desired to compare pressures which have been measured by means of different plasmolyzing solutions, it becomes necessary to make calculations which involve the molecular weights of the substances used. A much better way to express turgor pressures is in terms of fractions (e. g., tenths) of a molecular solution. But this, although it suffices for non-electrolytes, fails utterly for electrolytes, because of the unequal dissociation of different compounds. A y1^- gram- molecular solution of KNO3 will give a much greater osmotic 1 A. WLADIMIROFF, " Osmotische Versuche an lebenden Bakterien," Zeitschr.f. physik. Chem., Vol. VII (1891), pp. 529-43; also Zeitschr.f. Hygien, Vol. X (1891), pp. 89-110. TUEGIDITY 59 pressure than a ^ gram-molecular solution of glucose. A method must therefore be devised which will render it possi- ble to compare readily the osmotic pressures of electrolytes and non-electrolytes. This can be done by means of any unit of pressure. The mercury column may be used, or large pressures may be expressed in atmospheres. Recently Errera1 has suggested a special unit for measuring osmotic pressure, which he proposes to call the tonie. It is to be equal to the pressure of one dyne upon a surface of one square centimeter. For larger measurements he suggests the term myriotonie, equal to ten thousand tonies. It is difficult to see how this new unit possesses any advantage over the mercury unit for practical work. For plasmolytic purposes it is much more convenient to reduce all measure- ments to terms of a molecular solution of a non-electrolyte. Thus comparison becomes easy and the absolute pressure per unit surface can be readily found from the relation M= 22.3 atmospheres, or 1695 cm. Hg, where M is the osmotic pressure of a molecular solution of a non-electrolyte. Of course, in making up solutions of an electrolyte for use by this method it must be borne in mind that the desideratum is not a molecular solution of the elec- trolyte, but a solution whose osmotic pressure will just equal that of a given solution of a non-electrolyte. Thus a solu- tion of NaCl whose osmotic pressure is, say, T2T M, must be considerably more dilute than a T3F molecular solution of that salt. III. THE PERMEABILITY OF THE PEOTOPLASMIC LAYERS The often repeated statement that the protoplasmic layer is not permeable to solutes needs to be modified as follows: To some substances it is probably absolutely impermeable i L. ERRERA, " Sur la myriotonie comme unit6 dans les mesures osmotiques," Extr. des Bull, de Vacad. ray. de Belgique, Vol. Ill (1901), pp. 135-53. 60 DIFFUSION AND OSMOTIC PRESSURE under certain conditions; to the majority of substances it is usually very slightly permeable, but under certain conditions its permeability may increase ; and to some substances it is usually very readily permeable. Further than this, the pro- toplasm of different plants, and even of different cells in the same plant, has different osmotic properties. The condition of things is thus seen to be very complex. It will be of value to pass in review the most important fragments of evi- dence which have been accumulated upon this question of protoplasmic permeability. a) Test by the plasmolytic method. — There are several ways of testing the permeability of the protoplasmic sac. The one most frequently resorted to is that of plasmolysis. A bit of tissue or a unicellular organism is subjected to the osmotic action of solutions of the substance which is to be tested, these being of several different concentrations. If plasmolysis occurs in a solution of rather high concentration, this fact is taken as evidence that the protoplasm of the given cells is either impermeable to the solute or very slightly permeable. Of course, it is also theoretically pos- sible that in this case the substance used penetrates the protoplast to some extent and causes a polymerization or precipitation of the osmotically active solutes within the sap. There is no evidence for this phenomenon, however, and its general improbability throws it out of the category of seri- ous objections to the plasmolytic method. If, after being left a short time in the plasmolyzing solution, the cells regain their normal condition, it shows either that the pro- toplasm is somewhat slowly penetrated, or else that some osmotic material has been secreted within the cell. If plas- molysis occurs at a very low concentration, it is sufficient proof that the substance enters the protoplasm; for such plasmolysis is due to alteration in the membrane through poisonous action, or to a precipitation or some similar change TURGIDITY 61 within the vacuole, either of which phenomena could not take place without penetration. If plasmolysis does not occur even at high concentrations, we have evidence that the pro- toplasmic sac is not only penetrable to the substance used, but that this substance has no marked immediate toxic action. Cane sugar, glucose, KNO3, and NaCl are usually found to produce permanent plasmolysis. Plant cells placed in concentrated solutions of these substances do not, as a rule, regain their original turgid condition as long as they remain therein ; no perceptible penetration occurs. However, there are many cells whose protoplasts are more or less permeable to these compounds, and there are all gradations between absolute impermeability and rather slow permeability. One extreme of this series is Massart's Bacterium termo,1 which was not plasmolyzed at all in strong solutions of cane sugar and KNO3. But in most cases plasmolysis is the first result of irri- gating the cells with the test solution, and it is only after the lapse of some time that the first effect disappears. The gradual inward diffusion of the external osmotic substance, or, in some cases, the gradual secretion of an osmotic substance within the cells, finally brings about an equalization of the internal and external pressures. Then the original internal pressure, produced by the solutes within the vacuole, becomes again effective in producing turgidity. De Vries2 found that the tonoplasts of various plant cells were gener- ally freely penetrated by acids and alkalies, but that salts passed these membranes much more slowly. However, many cells were found which, after being plasmolyzed in a solution of KNO3 or NaCl, gradually returned to their origi- i MASSART, "SensibilitS et adaption des organismes a la concentration des solutions salines," Arch, de biol., Vol. IX (1899),pp. 515-70. 2H. DE VRIES, " Plasmolytische Studien tiber die Wand der Vacuolen," Jahrb. f. wins. Bot., Vol XVI (1885), pp. 465-598. 62 DIFFUSION AND OSMOTIC PRESSURE nal condition if left in the plasmolyzing solution. This author also observed that the presence of an acid or base, or of any other poisonous substance, made the protoplasm rapidly permeable to such salts as KNO3 and NaCl. The cells of the epidermis of leaves of Tradeacantia, Curcuma, and Begonia rex appeared to be impermeable to KNO3. The same author1 found the protoplasm of beets to be per- meable to NaCl. Janse2 found a similar return of tur- gidity in the case of marine algae (e. g., Chaetomorpha) which were allowed to remain in a solution of KNO3 or of NaCl which plasmolyzed them at first. He also found that the protoplasts of these algae are permeable to cane sugar. When plasmolysis was brought about in a solution of this substance, turgor gradually returned, but this process took about four times as long here as in a KNO3 solution. In Spirogyra the same general facts were observed, but the permeability is not as marked here as in the marine forms. In summing up the results of his second paper, this author states that he has found the protoplasm of the following five plants permeable as follows : Cheetomorpha is permeable to KNO3, NaCl, cane sugar. Spirogyra is permeable to KNO>, NaCl, grape sugar. Tradescantia and Curcuma are permeable to KNO3, NaCl. Stratiotes is permeable to KNO3. Glycerin and urea have been shown by De Vries3 and Klebs4 to penetrate nearly all plant cells with great readi- i H. DE VRIES, " Sur la permeabilit6 du protoplasma des betteraves rouges," Arch, uteri., Vol. VI (1871), pp. 117-26. 2J. M. JANSE, " Plasmolytische Versuche an Algen," Bot. Centralbl., Vol. XXXII (1887), pp. 21-6; IDEM, "Die Permeabilitat des Protoplasma,1' Verslag. en Mededeel. d. k. Akad. v. Wetensch. te Amsterdam, 3 Reihe, Vol. IV (1888), p. 332. 3H. DE VRIES, "Ueber den isotonischen Coefficient des Glycerins," Bot. Zeitg., Vol. XLVI (1888), pp. 229 ff. ; IDEM, " Ueber die Permeabilitat der Protoplaste fur Harnstoff," ibid., Vol. XL VII (1889), pp. 309 ff. *G. KLEBS, "Beitrage zur Physiologie der Pflanzenzelle," Unters. aus d. hot. Inst. zu Tubingen, Vol. II (1888), pp. 489 ff. ; IDEM, " Beitrage zur Physiologie der Pflanzenzelle," Ber. d. deutsch. hot. Ges., Vol. V (1887), pp. 181-9. TUEGIDITY 63 ness. Plasmolysis occurs, but is of short duration ; Overton1 found that it took from two to five hours for equilibrium to be re-established in solutions of these substances. There are exceptions here also, however, for De Vries found that the cells of the bud scales of Begonia manicata were almost impermeable to glycerin and urea. Jennings2 states that paramoecia are permanently plasmolyzed in glycerin. By an extensive investigation of the plasmolytic behavior of various organic compounds Overton3 found that a great number of these do not produce plasmolysis at all, so rapidly do they penetrate the protoplasm. Among these substances may be named : ethyl alcohol, ethyl ether, formaldehyde, chloral hydrate, acetone, methyl cyanid, furfurol, caffein, etc. The list includes practically all of the aliphatic alco- hols and related compounds which are soluble in water, and also a number of soluble aromatic compounds, such as anilin, acetanilid, phenol, phloroglucin, etc. Although the author does not express himself on this point, it seems probable that the apparent plasmolysis which occurs when plant cells are placed in strong alcohol is due, not to osmotic pressure, but to an increased permeability in the osmotic membranes (due to the poison) and also to an active con- traction on the part of the protoplasm. The same author shows that there are all gradations in rapidity of penetra- tion, from those substances which fail to plasmolyze at all to those which produce permanent plasmolysis. In all these cases he found that the protoplasmic sac is as readily per- meated outward as inward. Practically all liquids which 1 E. OVEETON, " Ueber die osmotischen Eigenschaften der lebenden Pflanzen- und Tierzelle," Vierteljahrschr. d. Naturf.-Ges. in Zurich, Vol. XL (1895), pp. 159-84. 2 H. S. JENNINGS, " Studies on the Reactions to Stimuli in Unicellular Organ- isms": I, "Reactions to Chemical, Osmotic, and Mechanical Stimuli in the Ciliate Infusoria," Jour. PhysioL, Vol. XXI (1897), pp. 258-322. 3 E. OVEKTON, loc. cit. The table of compounds occurs on p. 181. 64 DIFFUSION AND OSMOTIC PRESSURE are soluble in water penetrate readily, glycerin being one of the slowest. It appears as if the power to penetrate decreased with increasing specific gravity. Overton finds that the same thing is generally true of animal cells also, and — what is still more striking — that the amount of alcohols, etc., which plant and animal cells are able to bear is nearly the same. When the solute fails to penetrate the protoplast, the osmotic concentration necessary for plasmolysis is constant, no matter what the solute may be. But the more readily it penetrates, the higher the concentration necessary to bring about plasmolysis, until at last, as in the alcohols and ethers, this phenomenon does not truly occur at all. 6) Direct test of penetrability. — Another method of determining the extent of permeability manifested by pro- toplasm is to identify the diffusing substance after it has passed the plasmic layer. De Vries1 showed that the pene- tration of dilute ammonia into the cells of the red beet can be demonstrated by the reaction of the colored cell sap to this substance. The red sap changes to blue upon contact with an alkali. By choosing other cells whose sap contains red and blue dissolved pigments, Pf effer 2 showed that not only ammonia, but also the caustic alkalies and alkaline car- bonates as well as acids (such as tartaric, phosphoric, and carbonic) pass very rapidly through plant protoplasm. We may consider that this at least proves that the H and OH ions penetrate. In some cases a precipitate may be produced within the vacuole by the reaction of the penetrating sub- stance with the materials of the cell sap. This is the case with caffein, antipyrin, and some others.3 !H. DE VRIES, "Sur la perm6abilit6 du protoplasme des betteraves rouges," Arch. n6erl., Vol. VI (1871), p. 124; IDEM, uSur la mort des cellules v6g6tales," ibid. (1871), pp. 245-95. 2W. PFEFFER, Osmotische Untersuchungen, Leipzig, 1877, p. 140. 3 PFEFFER-EWART, Physiology of Plants, 1900, p. 98. TURGIDITY 65 Another manner of carrying out the direct test is to place cells in the solution to be tested and, after sufficient time has elapsed, to treat them with a reagent which will penetrate and also give a visible reaction with the substance to be tested. Thus, if penetration took place in the first solution a microchemical test for the solute should be obtained within the vacuole. In this way diphenylamin was first used by Molisch1 to detect KNO3 in plant cells. With this reagent a nitrate is indicated by the appearance of a blue color. It appears from the work of Molisch, Wieler, De Vries, and Janse, in which this method was used, that plant protoplasm is very generally penetrated by KNO3 in dilute solution. In a similar manner tests have been made with Fehling's solution, which prove the penetration of glucose. Janse 2 showed in this manner that Spirogyra protoplasts are penetrated by KNO3. Wieler3 worked with entire plants of the angiosperm group and obtained similar results. By means of diphenylamin and sulphuric acid he was able to demonstrate that NO3 ions penetrate the protoplasts of seed- ling beans, sunflowers, etc. By platinum chlorid he also demonstrated the penetration of K ions. Stems placed in a sugar solution formed starch, while a control without sugar failed to do so. It made no difference whether cane sugar or glucose were used. This proves the power of these sugars to penetrate the protoplasts in stems. The last-named author also presents evidence that the roots of seedlings of Vicia faba are able to absorb glucose from a solution. Of 1 H. MOLISCH, " Ueber den mikrochemischen Nachweiss von Nitraten und Nitriten in der Pflanze mittelst Diphenylamin oder Brucin," Ber. d. deutsch. hot. Ges., Vol. I (1883), pp. 150-55 ; also Sitzungsber. d. kais. Akad. d. Wiss. zu Wien, math.- nat. hist. Klasse, Vol. I (1887), p. 221. 2 J. M. JANSE, " Plasmolytische Versuche an Algen," Bot. Centralbl., Vol. XXXII (1887), pp. 21-6; IDEM, "Die Permeabilitat des Protoplasma," Verslag. en Mededeel, d. k. Akad. v. Wetensch. te Amsterdam, 3 Reihe, Vol. IV (1888), p. 332. 3 A. WIELEE, "Plasmolytische Versuche mit unverletzten Phanerogamen," Ber. d. deutsch. bot. Ges., Vol. V (1887), pp. 375-80. 66 DIFFUSION AND OSMOTIC PRESSURE course it is possible that, in these cases, the substance in question does not pass the protoplasm as such, but is modified at the surface of the ectoplast and penetrates in another form. For this question there seems to be as yet no method of attack. Wortmann1 believed he had evidence that the starch in the endosperm of seeds was not acted upon by an enzyme but by the protoplasm itself ; this, however, has been disproved.2 Since protoplasm contains so many enzymes of one sort and another it seems impossible to gather evidence as to whether a given action takes place within the proto- plasmic mass or outside of it. It is probable that it occurs wherever the enzymes are present, whether this be within or without. The penetration of anilin dyes has been studied exten- sively by Pfeffer,3 who showed, for example, that the sap of living cells may be strongly stained by the inward diffusion of methyl blue, methyl violet, etc. Certain anilin dyes may be absorbed into the living protoplasm itself and held there so as to give a marked stain. Even the nucleus may be so stained while living by means of dahlia, mauvein, etc.* In most of these experiments there is an accumulation of the stain in the vacuole or within the protoplasm. Thus, if plants of Elodea canadensis be placed for several days in a weak solution of methyl blue, they become visibly stained, while the external solution loses its color. Examination shows that the protoplasm itself is not colored, but that the 1 J. WORTMANN, " Ueber den Nachweiss, das Vorkommen und die Bedeutung des diastatischen Enzyms in den Pflanzen," Bot. Zeitg., Vol. XLVIII (1890), pp. 581 ff. 2B. HANSTEEN, "Ueber die Ursachen der Entleerung der Reservestoffe aus Samen," Flora, Vol. LXXIX (1894), pp. 419-29. 3 PFEFFER-EWART, Physiology of Plants, Cambridge, 1900, p. 96. References to an extensive literature are there given, the most important of which is : W. PFEFFER, "Ueber Aufnahme von Anilinfarben in lebenden Zellen," Unters.ausd. bot.Inst.zu Tubingen, Vol. II (1886), pp. 179-331 ; see also E. OVERTON, " Studien ttber d. Aufnahme der Anilinfarben durch die lebende Zelle," Jahrb. f. wiss. Bot., Vol. XXXIV (1900), pp. 669-701. *D. H. CAMPBELL, "The Staining of Living Nuclei," Unters. aus d. bot. Inst. zu Tubingen, Vol. II (1888), pp. 569-81. TURGIDITY 67 dye has accumulated in the vacuole, there becoming much more concentrated than was the original external solution. This phenomenon will be discussed under e). Another method by which direct determinations of per- meability may be made is to analyze the plant or its juice after the culture has been grown some time in a medium of known content. In this way von Mayenburg1 found that, out of a series of substances used in his culture fluids for Aspergillus niger, only glycerin was absorbed in sufficient quantity to be worthy of consideration as an osmotically active solute within the cells. c) Absorption test. — If the concentration of the sur- rounding medium is carefully determined and the organisms whose permeability is to be tested be allowed to grow in it for a time, decrease in concentration of the medium may be interpreted to mean that absorption has taken place. This has been shown to occur in the case of a number of inorganic salts, but is especially well marked with solutions of glucose and glycerin. Demoussy2 determined in this way the relative rate of absorption of potassium and calcium ions by wheat, maize, etc., while Laurent3 was able to prove the somewhat unexpected fact, that roots of maize can absorb measurable quantities of glucose from a solution in which they are grown. This is probably not a general phenomenon, for if the protoplasm is permeable to sugar in one direction it is difficult to see how it could fail to be permeable in the oppo- site one also, and such a condition must allow outward diffusion of glucoses and other sugars which are found so commonly in plant cells.4 i O. H. VON MAYENBURG, " Losungsconcentration und Turgorregulation bei den Schimmelpilzen," Jahrb.f. wiss. Bot., Vol. XXXVI (1901), pp. 381-420. 2 E. DEMOUSSY, "Absorption Elective de quelques elements mineraux paries plantes," Compt. rend., Vol. CXXVII (1900), pp. 970-73. 3 J. LAUKENT, " Sur 1'absorption des matieres organiques par les racines," ibid., Vol. CXXV (1897), pp. 887-9. * C/. PFEFFEE-EWABT, Physiology of Plants, Cambridge, 1900, p. 99. 68 DIFFUSION AND OSMOTIC PRESSURE That ordinary leaves can absorb inorganic salts was shown in this way by Dandeno.1 He found that drops of solution placed upon foliage leaves were completely absorbed if too rapid evaporation was prevented. When the drop disap- peared no trace of solute crystals remained upon the Isaf surface. d) Test by toxicity. — To all protoplasmic poisons must be accredited power of penetration in a greater or less degree ; if there were no penetration the substance could not bring about its toxic effect. True2 proved a slight toxicity for KNO3 and NaCl upon Spirogyra; these substances must therefore penetrate the protoplasts of this plant, though probably this occurs with difficulty. More recently Coupin3 has prepared a catalogue of the poisonous effects upon wheat of certain salts in various concentrations. His tables are useful for comparison. Of course, the fact that a rather high concentration of a given solute is needed to affect the plant may mean either that the protoplasm is only slightly permeable, or that the substance is only slightly toxic. From Pfeffer* we have the fact that mercuric chlorid and iodin penetrate many vegetable cells and exert a marked toxic effect. There are many other proofs that various mineral and organic substances are able to penetrate the plant protoplast. Where a noticeable and specific effect is produced upon the organism by the presence of a given substance in the medium, there can be no doubt that the substance pene- i J. B. DANDENO, "An Investigation into the Effects of Water and Aqueous Solutions of Some of the Common Inorganic Substances on Foliage Leaves," Trans, Canad. Inst., Vol. VII (1901), pp. 238-350. 2R. H. TRUE, "The Physiological Action of Certain Plasmolyzing Agents," JBot. Gaz., Vol. XXXVI (1898), pp. 407-16. 3H. COUPIN, "Sur la toxicit6 des composed du sodium, du potassium, et de 1'ammonium & 1'egard des vegetaux superieurs," JRev. gen. hot., Vol. XII (1901), pp. 177-94. * W. PFEFFER, Osmotische Untersuchungen, Leipzig, 1877, p. 140. TURGIDITY 69 trates, to some extent at least. Whether the effect be the death of the plant or only an alteration in its metabolic processes makes no difference for the present consideration.1 e) Test by accumulation. — The power of penetration of all inorganic salts, and of many organic compounds also, may be tested by analysis of plant material which has been grown in the solution to be tested. The various metallic ions such as K, Na, Ca, etc., are known to accumulate in the bodies of higher plants. In this manner Bourget2 demon- strated a marked absorption of iodin by various plant roots. There is a great difference in different plants in this regard, however; the Liliaceae and Chenopodiaceae absorb com- paratively large quantities of iodin, while Solanum tuberosum, grown in the same soil, fails to absorb enough for a test. Great accumulation of copper in plant cells has been recorded several times. Thus, MacDougal3 describes a case where a tree of Quercus macrocarpa absorbed copper in large amounts and caused its precipitation within the wood in the metallic state. All these accumulations come about by a chemical change taking place in the substance after it has entered the cell, 1 The following references will be of use to supplement those already given : F. DE F. HEAJLD, " On the Toxic Effect of Dilute Solutions of Acids and Salts upon Plants," Bot. Gaz., Vol. XXII (1896), pp. 125-53; H. M. RICHARDS, "Die Beeinflussung des Wachsthums einiger Pilze durch chemische Reize," Jahrb.f. wiss. Bot., Vol. XXX (1897), pp. 665-79; F. L. STEVENS, "The Effect of Aqueous Solutions upon the Ger- mination of Fungus Spores," Bot. Gaz.,Vol. XXVI (1898), pp. 377-406; E. B. COPELAND AND L. KAHLENBERG, " The Influence of the Presence of Pure Metals upon Plants," Trans. Wisconsin Acad. Sci. Arts and Let., Vol. XII (1899), pp. 454-74; N. ONO, " Ueber die Wachsthumsbeschleunigung einiger Algen und Pilze durch chemische Reize," Jour. Coll. Sci. Imp. Univ. Tokyo, Vol. XIII (1900), reviewed in Bot. Gaz., Vol. XXX (1900), p. 422; H. DE VAUX, " De 1'absorption des poisons metalliques tres dilues par les cellules vegetales," Compt. rend., Vol. CXXXII (1901), pp. 717-20. 2 P. BOURGET, " Sur 1'absorption de 1'iode par les vegeteaux," Compt. rend., Vol. CXXIX (1899), pp. 768-70; IDEM, same title, Bull. soc. chim. Paris, Ser. 3, Vol. XXIII (1899), pp. 40-41. 3 D. T. MACDOUGAL, " Copper in Plants," Bot. Gaz., Vol. XXVII (1899), p. 68. He cites the following on the same general subject : LEHMAN, "Der Kupf ergehalt von Pflanzen und Thieren in kupferreichen Gegenden," Arch. f. Hygien, Vol. XXVII (1896), p. 1; J. B. SKERTSCHLT, "Tin Mines of Watsonville," Report Geologist Queensland, 1897. 70 DIFFUSION AND OSMOTIC PRESSURE either within the protoplasm or in the vacuole. If this were not so, the diffusion tension of the solute would soon become as great within the cell as without, and thus there could be no accumulation. But if a substance is precipitated, poly- merized, or condensed within the cell through the chemical action of some other body already there, which perhaps arises as a secretion from the protoplasm, then the internal diffusion tension of the entering substance will be kept low, and inward diffusion will continue indefinitely. In this way copper salts entering the cell are probably reduced to metallic copper. This fact of accumulation is a very important one in under- standing the process of absorption of dissolved substances by the plant. f) Test by metabolic processes. — The absorption of any food substance is of course a proof of permeability to that substance. The immediate effect upon the living green cell of absence of carbon dioxid, or upon any living cell of oxygen, shows that these gases, when in solution, enter the protoplast with extreme ease. Penetration by many usually solid substances may be proved in this manner ; the long series of experiments upon growth, and especially the forma- tion of starch by green plants in darkness, may be regarded as evidence in this matter. Thus, Bouilhac1 grew Nostoc in the dark in a solution of glucose, where it appeared perfectly healthy, and Artari2 and Matruchot and Molliard3 grew Stichococcus in organic solutions in a similar way. The absorption of organic food by many algae, and by all saporo- phytes and parasites, including all of the fungi, may be mentioned in this connection. In many of these cases the 1 R. BOUILHAC, "Sur la culture de Nostoc punctiforme en presence de glucose," Compt. rend., Vol. CXXV (1897), p. 880. 2 A. ARTARI, "Zur Ernahrungsphysiologie der grunen Algen," Ber. d. deutsch. bot. Ges., Vol. XIX (1901), pp. 7-10. 3 L. MATRUCHOT ET M. MOLLIARD, " Variations de structure d'une algue verte sous 1'influence du milieu nutritif," Rev. gen. bot., Vol. XL (1902), pp. 114-30, 254-68, TURGIDITY 71 presence of the substance in question is due to digestion outside the body, brought about by outward diffusion of enzymes. Dandeno1 has recently shown that inorganic salts are absorbed in some instances by ordinary leaves when these are kept covered by a solution by means of a constant spray, or by submersion. This occurred to such a degree in this writer's experiments with Thunbergia that plants of this form whose roots were supplied with nothing but water, but whose leaves were sprayed with a solution, were able to make a good growth. Control plants, which had distilled water applied to the leaves as well as to the roots, perished in a much shorter time. Also, drops of solution placed upon various leaves were completely absorbed if too rapid evapora- tion was prevented. This observation has been mentioned under c). g) Outward permeability. — Many substances which pene- trate the cell from without have been shown to pass in the opposite direction with equal ease. This has been especially emphasized by Overton 2 in the case of the soluble alcohols, etc. But in general the outward passage from the plant body of sugars and the various organic food substances has not been demonstrated. It must be of rather rare occurrence, or the phenomena of nutrition, etc., would be impossible. There are, however, certain cases where exudation occurs, notably in the case of glandular structures, both in plants and animals. Laurent,3 however, has demonstrated an out- ward passage of enzymes (e. g., amylase and sucrase) from 1 J. B. DANDENO, "An Investigation into the Effects of Water and Aqueous Solutions of Some of the Common Inorganic Substances on Foliage Leaves," Trans. Canad. Inst., Vol. VII (1901), pp. 238-350. 2 E. OVEETON, " Ueber die osmotischen Eigenschaften der lebenden Pflanzen und Thierzelle," Vierteljahrschr. der Naturf.-Ges. in Zurich, Vol. XL, (1895), pp. 159-84. 3 J. LAURENT, " Sur Texosmose de diastases par les plantules," Compt. rend., Vol. CXXXI (1900), pp. 848-51. 72 DIFFUSION AND OSMOTIC PRESSURE the roots of maize seedlings, and a similar phenomenon is very commonly met with in the case of bacteria, yeast fungi, etc. Molisch l believed this to be generally true, but Czapek 2 has shown that he was probably mistaken. The latter found that normal roots give off not only CO2 but also phosphoric acid in the form of an acid salt. These substances must of course pass out through the protoplasm. There seems to be no doubt from the work of Dandeno3 that both organic and inorganic substances will diffuse out from the cells of foliage leaves if these are kept covered with water. In these cases inward and outward diffusion seem to take place in exactly the same manner. It is a well-known fact that enzymes, especially diastase, pass out from the cells of embryos and digest food stored in the endosperm of seeds.4 This argues the permeability of the protoplasm of both embryo and endosperm to these substances. In this connection evidence has also been presented that the cells of embryo and endosperm are both permeable to carbohydrates, probably of the glucose group. Whether these arise from the action of an enzyme derived from the embryo, or from the action of enzymes formed within the endosperm itself,5 is of no consequence so far as permeability is concerned; after the carbohydrates are formed they diffuse into the embryo. h) Variations in permeability. — If a turgid cell gives 1 H. MOLISCH, " Ueber Wurzelausscheidungen und deren Einwirkung auf orga- nische Substanzen," Sitzungsber. d. kais. Akad. d. Wiss. zu Wien, math.-nat. hist. Klasse, Vol. XCVI (1887), pp. 84-109. 2F. CZAPEK, "Zur Lehre von den Wurzelausscheidungen," Jahrb.f. wiss. Bot., Vol. XXIX, pp. 321-90. 3Loc.cit.,p.ll. *GBUSS, " Ueber d. Eintritt von Diastase in d. Endosperm.," Ber. d. deutsch. bot. Ges. zu Berlin, Vol. IV (1893), p. 286; also B. HANSTEEN, "Ueber die Ursachen der Entleerung der Reservestoffe aus Samen," Flora, Vol. LXXIX (1894), pp. 419-29. 5 K. PURIEWITCH, "Ueber die selbstthatige Entleerung der Reservestoffbehalter," Ber. d. deutsch. bot. Ges., Vol. XIV (1896), pp. 207-15; IDEM, " Physiologische Unter- suchungen fiber die Entleerung der Reservestoffbehalter," Jahrb. f. wiss. Bot., Vol. XXXI (1897), pp. 1-76. TUBGIDITT 73 out pure water, this occurrence must be due to one of two conditions : (1) the protoplasm may contract with great force, thus overcoming the osmotic pressure of the contained solutes and causing the solvent to pass outward through the membrane, or (2) a relative decrease in the internal pressure may occur resulting either from an active precipita- tion or condensation of some of the solutes of the sap, or from an absolute rise in the external osmotic pressure. Accord- ing to supposition (1), the osmotic pressure within the cell remains unchanged, but is in part overcome by the mechani- cal pressure of the contracting protoplasmic membrane. According to supposition (2), the internal osmotic pressure is relatively reduced, and the protoplasm does not exert any appreciable pressure itself, but is forced inward through the solvent by the osmotic pressure of the solutes outside the cell and by the elastic force of the restraining cellulose wall. It is probable that this last supposition expresses the truth in many cases where an alteration in turgidity is observed. The former supposition is not tenable at all ; the protoplast would burst long before concentration of the sap solution could be brought about by pressure. If a cell gives out a solution, the cause of this must be a change in the permeability of the protoplasm, such that it now allows the outward passage of solutes to which it was formerly impermeable. The liquid exuded in guttation is known1 to be, not pure water, but a portion of the cell sap. In the last particular this sort of shrinkage of the vacuole differs from true plasmolysis, for in that we have the extrac- tion of pure water. However, the apparent effect upon the cell is the same ; if the volume of the vacuole is in any way decreased, the protoplasmic sac will contract from its own elasticity and surface tension, if for no other reason. 1 G. BONNIEE, " Recherches exp6rimentales sur la miellee," Rev. gen. hot., Vol. VIII (1896), pp. 1-22; DANDENO has shown that guttation droplets contain both organic and inorganic solutes. See Trans. Canad. Inst., Vol. VII (1901), pp. 238-350. 74 DIFFUSION AND OSMOTIC PRESSURE At different times and under different conditions the permeability of certain protoplasts apparently changes greatly. The presence of poisons may cause the protoplasm to become more permeable to other substances. Thus, Maquenne1 found that HgCl2 caused a marked increase in the permeability of the protoplasm of the cells of Helianthus seedlings to plasmolyzing agents. Similarly DeVries2 found that plasmatic membranes which were normally impermeable to KNO3 and NaCl could often be made permeable to them by treatment with an acid or a base. With animal muscle Loeb3 has shown that acids, bases, and other chemicals exert a great influence upon the water- absorbing power of the cells. This may be due to changes in permeability. Partial or complete plasmolysis may act in the same way. Oltmanns * was able to cause Fucus cells to give out coloring matter by placing the tissues in concentrated solutions. Both of these reactions must consist in an alteration of the physical (perhaps chemical) structure of the protoplasm. On the other hand, an increase in turgor above the normal may cause the same change. This seems to be the case in the guttation from the water pores of the leaves of the tomato, balsam, etc. When the turgor pressure in the cells bordering these water pores passes a certain limit, the protoplasm apparently becomes altered so that the cell sap oozes out and appears in droplets on the leaf-tips where water pores are present. Czapek5 describes a similar phe- i L. MAQUENNE, " Sur la pression osmotique dans les graines germees," Compt. rend., Vol. CXXIII (1896), pp. 898, 899. 2H. DE VEIES, " Plasmolytische Studienft. d. Wand d. Vacuolen," Jahrb.f. wiss. BoL, Vol. XVI (1885), pp. 465-598. 3J. LOEB, " Physiologische Untersuchungen fiber lonenwirkung " : I. Mitthei- lung, " Versucheam Muskel," Pflilgers Arch.f. d. ges. PhysioL, Vol. LXIX (1897), pp. 1-27. *F. OLTMANNS, "Ueber die Bedeutung der Concentrationsanderung des Meerwassers fur das Leben der Algen," Sitzungsber. d. k. preitss. Akad. d. Wiss. zu Berlin, Vol. X (1891), p. 183. 5 F. CZAPEK, " Zur Lehre von den Wurzelausscheidungen," Jahrb. /. wiss. Bot., Vol. XXIX (1896), pp. 321-90. TUEGIDITY 75 nomenon in the case of turgid root hairs, from which droplets of solution are exuded. It is probable that in these cases we have to do with a phenomenon related to glandular secretion. Another potent cause for great increase in protoplasmic permeability in some instances is lowering of temperature. If a filament of any common alga be carefully dried exter- nally and placed in olive oil whose temperature is then rapidly lowered to the vicinity of 0° C., a film of water may be seen to form about the filament, and partial plasmolysis may be observed. When the temperature is again brought back to normal, the extruded water is again absorbed. Greeley l has recently shown, not only that complete plas- molysis can be produced in Spirogyra by low temperature, but that the same thing occurs in Stentor coeruleus. Exactly the same phenomenon is exhibited by Stentor individuals when water is removed from them by the action of a con- centrated sugar solution. The animals plasmolyzed by low temperature return to their normal activity with rise in tem- perature, but Greeley was unable to cause the same reversal in the case of the osmotically plasmolyzed individuals. I have often observed that the liquid exuded from cells of Spirogyra plasmolyzed by cold is a solution. Its freezing point is considerably lower than that of pure water. The theory of death by freezing which was advanced by Molisch2 accounts for the decline of activity and for final death at low temperatures by the extraction of water from the protoplasm until the processes which make up life are no longer possible. Matruchot and Molliard3 have pointed i A. W. GREELETT, " On the Analogy between the Effects of Loss of Water and Lowering of Temperature," Am. Jour. PhysioL, Vol. VI (1901), pp. 112-28. 2H. MOLISCH, Untersuchungen ilber das Erfrierender Pflanzen, Jena, 1897; reviewed in Bot. Centralbl., Vol. LXXIII (1898), p. 149. 3L. MATROCHOT AND M. MOLLIARD, "Sur Tidentite des modifications de structure produites dans les cellules vegetales par le gel, la plasmolyse, et la fanaison," Compt. rend., Vol. CXXXII (1901), pp. 495-8. 76 DIFFUSION AND OSMOTIC PRESSURE out a striking parallelism in the behavior of plant nuclei which have been either frozen, dried, or subjected to the osmotic action of a concentrated solution. In all these cases water was found to be extruded from the nucleus. The nuclear material took on a peculiar appearance not unlike that of karyokinetic figures. Krabbe 1 experimented upon the effect of rise in tempera- ture upon the absorption of water by various plant cells, finding that the rate of absorption rises with the tempera- ture. This author supposes the response to be due to a physical change in the protoplasm, caused by the higher temperature. But the best series of experiments on the change in pro- toplasmic permeability due to temperature variations is that of van Kysselberghe.2 He worked with a variety of plant cells (Sambucus, Tradescantia, Begonia, Lemna, green algaB, etc.) and found that Krabbe's general result is true. The ratio of increase in permeability to water becomes less, however, as the temperature rises. From 0° C. to 5° C., this ratio is 0.05 ; from 5° C.to 18° C., 0.043 ; and above the last-named tempera- ture, 0.1. The ratios between the permeability to water at 0° and that at 6°, 12°, 16°, 20°, 25°, 30° are: 1, 2, 4.5, 6, 7, 7.5, 8. The total amount of water absorbed by a cell is not changed by variations in temperature ; the rate of absorption alone is affected. The nature of the protoplasm does not appear to have any effect on the total amount of water absorbed or given out by a cell ; this is determined by the osmotic pres- sure of the sap and by the temperature. Permeability to !G. KRABBE, "Ueber d. Einfluss d. Temperatur auf d. osmotische Processe lebender Zellen," Jahrb.f. wiss. Bot., Vol. XXIX (1896), pp. 441-98. 2F. VAN RYSSELBERGHE, "Influence de la temperature sur la perm6abilit6 du protoplasme vivant pour 1'eau et les substances dissoutes," Recueil de Vinst. bot. de Bruxelles, Vol. V (1901), pp. 209-49; IDEM, " Reaction osmotique des cellules v6getales a la concentration du milieu," Mem. cour. pub. par Vacad. roy. de Selg., Vol. LVIII (1898), pp. 1-101. TUEGIDITY 77 this liquid does not cease altogether at 0° C., as was thought by Schwendener.1 At 0° C. protoplasm is not only perme- able to water, but also to KNO3, glycerin, urea, methylene blue, caffein, and ammonium carbonate. Thus Krabbe's idea that below 5° 0. nothing but water penetrates is entirely unfounded. Temperature variations in permeability to solutes were also observed in many cells by the same author. These, too, seem to follow the rule for water as stated on the preceding page. However, Copeland2 showed that decrease in temperature caused a rise in tde turgor pressure of moss leaves. This may not be a direct effect of the temperature upon the pro- toplast, for the same author found that various agencies which checked growth also caused a rise in turgor ; there is surely a close relation between growth and turgor. what- ever this relation may ultimately turn out to be. In this connection it may be noted that De Vries3 found that, as growth proceeds, turgor rises, to fall again after the curve of growth begins to decline. The extrusion of liquid from the cells of the pulvini of "sensitive" organs, such as the leaves of Mimosa and the stamens of Berberis, may be due to a change in permeability also. There seems to be a question, however, as to whether the exuded liquid is water or a solution. Pfeffer4 considers this subject at some length, and concludes that solutes prob- ably do not pass out. He believes that the salts found by Janse5 in the extruded liquid from pulvini in Mimosa are 1 S. SCHWENDENER, " Zur Kritik der neuesten Untersuchungen ft. d. Saftsteigen," Sitzungsber. d. k. preuss. Akad. d. Wiss. zu Berlin, Vol. VI (1892), p. 911. 2 E. B. COPELAND, Ueber den Einfluss von Licht u. Temperatur auf den Turgor, Halle a. S., 1896. 3H. DE VRIES, "Ueber die Ausdehnung wachsender Pflanzenzellen durch ihren Turgor," Bot. Zeitg., Vol. XXXV (1877), p. 1. s . Thus, the internal and external osmotic pressure will be more nearly the same at a low temperature than at a higher one. The two pressures should become equal at absolute zero. No measurements have been made to determine whether the decrease in volume of the Spirogyra vacuole is propor- tional to the approach of the external and internal concen- trations toward each other. This should not be a difficult thing to settle. But, as has already been stated (page 75), there is cryoscopic evidence that the extruded liquid is not pure water. The identity of the responses obtained by Loeb with Copepods and Polygordius larvae when these were subjected to cold and to high concentrations, has also been noted (page 139). A similar change of tropism occurs among those plant lice which exist in two forms, one winged and the other wingless. The growth of wings in the wingless form can be called forth either by low temperature or by allowing the plants upon which the animals are feeding to dry, thus depriving the latter of water. While in the wingless condition these lice are negatively heliotropic, but upon devel- oping wings they become positively so. Here is a reversal of tropism brought about by withdrawal of water, but this experiment also shows that, although the general protoplas- mic activity may be depressed by this treatment, yet certain special activities (e. g., those involved in wing formation) may be accelerated. INFLUENCE OF THE MEDIUM 143 It is generally known that lowering of the temperature of an animal heart causes the beating to become less rapid. This is perfectly parallel to the falling off in heart activity in strong solutions, as observed by Miss Shively and recorded by Loeb (page 129). In my own experiments on Stigeoclonium,1 it was found that the organism responds to drying on a porous plate in exactly the same way as it does to change from a weak to a strong solution. Recently, Greeley2 has shown that by cooling Stentor ccerulcus the same cessation of activity and rounding up was brought about as when the animals were subjected to the action of concentrated solutions. However, the effect of the solution was not reversible, for the animals could not be revived. The same author has shown that cold plasmolysis in Spirogyra is reversible, that a rise in temperature brings the plasmolyzed alga back to its normal condition. During the summer of 1901 Greeley3 was able to pro- duce artificial parthenogenesis of Echinoderm eggs by merely keeping them for a time at a low temperature. In these cold-fertilized eggs, development went as far as in normally fertilized ones under artificial conditions. In general, then, it may be concluded that there is a striking analogy between the responses obtained in these various organisms by treating them with strong solutions and by extracting water from them in any other way. How much further we may go in this, remains for future experi- ment to show. 1 B. E. LIVINGSTON, " Further Notes on the Physiology of Polymorphism in Green Algse," Bot. Gaz., Vol. XXXII (1901), pp. 292-302. 2 A. W. GEEELET, " On the Analogy between the Effects of Loss of Water and Lowering of Temperature," Am. Jour. Physiol., Vol. VI (1901), pp. 122-8. 3 A. W. GREELEY, "Artificial Parthenogenesis Produced by a Lowering of the Temperature," Am. Jour. Physiol., Vol. VI (1902), pp. 296-304. 144 DIFFUSION AND OSMOTIC PRESSURE III. SUMMARY OF THE CHAPTER As far as investigation has gone, it has been found that growth is accelerated in weak solutions and retarded in con- centrated ones. The term "growth" here includes, not only enlargement, but also the process of cell division. Also, in some cases at least, the direction of new walls is profoundly influenced by the concentration of the surrounding medium. In general, all vital processes are retarded in concentrated solutions. ^Reproduction, being a peculiar form of cell divis- ion, appears in some cases to be entirely dependent upon the osmotic pressure of the surrounding medium. Irritability is also greatly influenced by external pressure. Not only is this function retarded in concentrated solutions, but in some forms the direction of response to a given stimulus may be reversed by a sudden change in the osmotic surroundings. The comparative concentration of the external and internal solutions acts, in many cases, as a stimulus upon the organ- ism, giving rise to the phenomena of osmotaxis. All the effects of high concentration of the surrounding liquid seem to be due to extraction of water from the living cells. They may be due either to a drying-out process or to decrease in turgidity. That they are sometimes due to the former is proved by curious analogies between the vari- ous processes which extract water from the protoplasm. Whether or not this extraction of water from the protoplasm itself is the direct cause of the responses to concentrated solutions, is not yet known. The effect may be a chemical one, due to the increased concentration of the contained solutions. INDEX ABSORPTION : of gases, 115 ; of solids and liquids, 118; of solutes, 115. ACIDS : influence of, on permeability, 61, 74; penetrating power of, 64. ACETANILID, in plasmolysis, 63. ACETONE, in plasmolysis, 63. ACTION OF PROTOPLASMIC MEMBRANE, 80. ALCOHOL., ethyl, in plasmolysis, 63. ALCOHOLS: penetrating power of, 71; aliphatic, in plasmolysis, 63. ALKALIES, penetrating power of, 64. AMMONIA, penetrating power of, 64. AMMONIUM CARBONATE, 77. AMMONIUM CHLORIC, ionization of, 23. AMPHITRITE, eggs of, 130. AMYLASE, penetrating power of, 71. ANESTHETICS, effect of, on permeability, 78. ANILIN, in plasmolysis, 63. ANILIN DYES, penetrating power of, 66. ANIMAL CELLS, permeability of, 64. ANTIPYRIN, penetrating power of, 64. APPLE, pressure of sap, 86. ARBACIA, eggs of, 128. ARENGO, exudation pressure of, 102. ARRHENIUS, 18, 24. ARTARI, 70. ASKENASY, 111. ASPERGILLUS: permeability of, 67; so- lutes of, 83. ASTERIAS, eggs of, 129, 130. ASCI, bursting of, 54. ATMOSPHERE, internal, 120. ATOMIC THEORY, 3. AVOGADRO, principle of, 11. BACTERIA, plasmolysis of, 58, 61. BACTERIUM TERMO, plasmolysis of, 61. BASES, influence of, on permeability, 61, BASIDIOBOLUS, in^osmotic solutions, 132, 138. BEAN, permeability of, 65. BEAUVERIE, 137. BECKMANN, 37, 39. BEET: permeability of, 62, 64, 78; solutes of, 83. BEGONIA, permeability of, 62, 73, 76. BERBERIS, pulvini of stamens in, 76. BLACKMAN, 116, 117. BLEEDING, 102; theory of, 104. BLOOD CORPUSCLES, 54, 57. BLUE, methyl, penetrating power of, 66, 77. BOHM, 111. BONNIER, 73, 98. BOUILHAC, 70. BOURGET, 69, 119. BOWER, 88. BOYLE, principle of, 10. BROWNE AND ESCOMBE, 116. BUPFUM AND SLOSSON, 113. BURGARSZKY, 18. BURSTING OF CELLS, 54. CABBAGE, pressure of sap, 86. CAFFEIN : in plasmolysis, 63 ; penetrating power, 64, 77. CALCIUM: absorption of, 67, 120; pene- trating power of, 69. CALCIUM NITRATE, absorption of, 120. CAMPBELL, 66. CAOUTCHOUC, membrane of, 82. CARBON DIOXID : absorption of, 93, 115 ; from roots, 72 ; influence of, on trans- piration, 113 ; penetrating power of, 70. CARBONATES, penetrating power of, 64. CARROT, pressure of sap, 86. CELERY, pressure of sap, 86. CELL WALL, permeability of, 55. CHAETOMORPHA, permeability of, 62. CHAETOPTERUS, chemical fertilization of, 130. CHEMICAL THEORY OF SEMI-PERMEABIL- ITY, 82. CHENOPODIACEAE, absorption of iodin by, 69. CHLORIDS, penetrating power of, 78. CHROMULINA, reversal of tropism in, 140. CHLORAL HYDRATE, in plasmolysis, 63. CHOLESTERIN, 81. CILIA, in osmotic solutions, 139. CLAUSEN, 116. Cocos, exudation pressure of, 102. CODIUM, permeability of, 78. COEFFICIENTS, isosmotic, 56. COHNHEIM, 83. COLLOIDS, 27, 49. CONDUCTIVITY OFJSAPS, 85. COPELAND, 77, 83, 88, 105, 108, 111. COPELAND AND KAHLENBERQ, 69. COPEPODS, reversal of tropism in, 139. 145 146 DIFFUSION AND OSMOTIC PRESSURE COPPEE, accumulation of, 69. COPPEE FEEEOCYANID MEMBEANE, 82, 112. COPPEE SULFATE, 110. COUPIN, 68. CEYSTALLOIDS, 49. CUECUMA, permeability of, 62. CUETIS, 54. CUEVATUEE, rOle of turgidity in, 89. CYNAEA, 78. CZAPEK, 72, 74. DANDENO, 22, 68, 71, 72, 98. DAVENPOET, 31, 128. DEATH, theory of, 75. DEMOUSSY, 67, 119. DEEO VAOA, regeneration of, 128. DEVAUX, 69. DE VEIES, 55, 56, 61, 62, 64, 65, 74, 77, 83, 84. DIFFUSION, of gases, 9. DIFFUSION TENSION, of solvent, 30. DIGESTION, outside the body, 71, 72. DIXON, 111, 112. DUTEOCHET, 111. DYES, anilin, penetrating power of, 66. ECTOPLAST, 51, 53, 80, 82. EGGS, parthenogenesis of, by cold, 143. ELECTEOLYTES, in plasmolysis, 56. ELODEA, permeability of, 66. EMBEYO, permeability of, 72. ENDOSPEEM, permeability of, 72. ENGELMANN, 139. ENVIEONMENTAL FACTOES, 47. ENZYMES, penetrating power of, 71, 72. EPIDERMIS, permeability of, 117. ESCHENHAGEN, 131. ETHEE, ethyl, in plasmolysis, 63. EEEEEA, 59. EUPHOEBIA, nectaries of, 79. EVAPOEATION, 110. EXUDATION : nature of, 73; from wounds, 102; from glands, 71, 96. EXUDATION PEESSUEE, theory of, 104. FAGOPYEUM, solutes of, 83. FEHLING'S SOLUTION, 65. FEEEIC PYEOLIGNATE, 110. FEETILIZATION BY COLD, 143. FICK, 18. FlLTEE THEOEY OF SEMIPEEMEABILITY, 80. FLUSIN, 82. FOEM, retention of, 87. FOEMALDEHYDE, 63. FEAZEUE, 128. FUCHS, 104. Fucus, permeability of, 74. FUNGI, in osmotic solutions, 138. FUEFUEOL, 63. GASES: diffusion tension of, 9; mixed, 11 ; absorption of, 115 ; transmission of, 120. GAY-LUSSAC, principle of, 10. GLYCEEIN, in plasmolysis, 56, 62, 63, 64, 67, 77, 79. GLUCOSE: in turgor, 84; penetrating power of, 61, 65, 67, 70, 79. GODLEWSKI, 108. GOGOEZA AND GONZALEZ, 54. GEAHAM, 18. GEAM MOLECULE, 20. GEEELEY, 75, 143. GEOWTH, rOle of turgidity in, 88. GEYNS, 57. GUNNEEA, solutes of, 83. GUTTATION, 73, 74, 98. HAMBUEGEE, 57, 83. HANSTEEN, 66. HAETIG, 110. HAUPT, 79, 100. HEAETS : in osmotic solutions, 129 ; effect of cold upon, 143. HEALD, 69, 85. HEDIN, 57, 83. HELIANTHUS: permeability of, 65, 74; solutes of, 83, 84. HILBUEG, 78. HOBEE, 18, 83. HONEY-DEW, 98. IMBIBITION, 93. INFUSOEIA, in osmotic solutions, 132. INTEENAL ATMOSPHEEE, 120. IODIN, penetrating power of, 68, 69. IONS, in absorption, 53. IONIZATION: of gases, 23; of solutes in liquids, 24. IEEITABILITY, changes of, 139. JAEIUS, 130. JANSE, 62, 65, 77, 108. JENNINGS, 63. JUMELLE, 112. JUNG, 128. KAHLENBEEG, 22. KINETIC THEOEY OF MATTER, 4. KLEBS, 62, 88. KNOP'S NUTEIENT SOLUTION, 132. KOHLEAUSCH, 42. KOHLEAUSCH AND HOLBOEN, 42, 43. KOPPE, 57. KOSSAEOFF, 113. KOVESI, 83. KEABBE, 75. KEAUS, C., 102. KEAUS, G., 83. INDEX 147 LAURENT, 67, 71. LEAVES: absorption by, 68, 103; exuda- tion from, 99; guttation or, 74; permea- bility of, 71. LECETHINS, 81. LEHMAN, 69. LEITZMANN, 116. LEMNA, permeability of, 76. LICE, reversal of tropism in, 141. LIDFOESS, 54. LlEBEEMANN, 18. LlLIACEJS, 69. LIQUIDS: absorption of, 118; diffusion of, 12, 13. LIVINGSTON, 132, 138, 143. LOB, 57. LOEB, 53, 74, 128, 129, 139. LOEB, FISCHER, AND NEILSON, 130. LUPINUS: in osmotic solutions, 130; so- lutes of, 84. MACDOUGAL, 69, 119. MAQUENNE, 74, 84, 85. MASSART, 61, 140, 141. MATEUCHOT AND MOLLIARD, 70, 75. MATTER : nature of, 3 ; states of, 6. MATER, 84. MEDIUM, influence of, 124. MEERBUEG, 36, 82. MEMBEANES : cellulose, 51 ; copper ferro- cyanid, 82. MEMBEANES: protoplasmic, 49; action of, 80. MERCURIC CHLORID: effect on permea- bility, 74 ; penetrating power, 68. METHYL BLUE, penetrating power of, 66, 77. METHYL CYANID, in plasmolysis, 63. METHYL VIOLET, penetrating power of, 66. MIMOSA, pulvini of, 77. MOHL, 98. MOLDS, bursting of, 54. MOLISCH, 65, 72, 75, 102, 103, 106. MORGAN, 129. MOESE AND HOEN, 36. MUCOE, exudation from, 99. MULLER, 116. MUSCLE, permeability of, 74. MYEIOTONIE, 59. NACCARI, 18. NATHANSOHN, 78. NECTARIES : artificial, 99 ; conditions for secretion of, 101 ; theory of, 96. NERNST, 18. NERNST-PALMER, 37, 38, 40. NITRATES, test for, 65. NOLL, 54. NOSTOC, permeability of, 70. NUCLEI: frozen, dried, etc., 76; in osmo- tic solutions, 128. OLTMANNS, 74, 102. ONION : permeability of, 78 ; solutes of, 83. ONO, 69, 125. OSMOTAXIS, 140. OSMOTIC PEESSUEE: in general, 28; de- monstration of, 32; of electrolytes, 27; of non-electrolytes, 25 ; indirect meas- urement, by freezing-point, 37 ; by boil- ing-point, 38; by vapor tension, 39; calculation of, for electrolytes, 42; for non-electrolytes, 41; compared to dry- ing, 141. OSTWALD, 42. OSTWALD-WALKER, 16, 40, 42. OVEETON, 63, 71, 81. OXYGEN: absorption of, 115; effect of, on permeability, 78; penetrating power of, 70. PAEAMCEClA,?permeability of, 63. PARTHENOGENESIS : 129 ; by cold, 143. PEAR, pressure of sap of, 86. PEAS, solutes of, 84. PENICILLIUM, in osmotic solutions, 131. PERMEABILITY OF PROTOPLASM: 60, 64, 67, 68, 69, 70, 118; outward, 63, 71; varia- tions in, 72 ; effect of on turgidity, 86. PFEFFER, 35, 50, 55, 64, 66, 68, 77, 78, 104, 105, 120. PHASEOLUS: in osmotic solutions, 130, 138 ; permeability of, 65, 78 ; solutes of, 83. PHENOL, in plasmolysis, 63. PHLOROGLUCIN, in plasmolysis, 63. PHOSPHORIC ACID, from roots, 72. PHOTOSYNTHESIS, 117. PHYTOLOCCA, penetrating power of sap of, 98. PICRIC ACID, 110. PILOBOLUS, bursting of, 54. PISUM: in osmotic solutions, 130, 138; solutes of, 83, 84. PITRA, 103. PLANT LICE, reversal of tropism in, 142. PLASMOLYSIS : in general, 54, 60 ; by cold, 75 ; of bacteria, 58 ; effect of on growth, 88 ; on permeability, 74. PLATINUM CHLORID, 65. POISONS, permeating power of, 68. POLLEN GEAINS, bursting of, 54. POLYGOEDIUS, reversal of tropism in, 139. POLIMOEPHISM : in Basidiobolus, 137 ; in fungi, 138 ; in Stigeoclonium, 132. POTASSIUM : absorption of, 67, 120 ; pene- trating power of, 69. POTASSIUM CHLOEID, in turgor, 83. 148 DIFFUSION AND OSMOTIC PRESSURE POTASSIUM NITRATE : absorption of, 120 ; in plasmolysis, 56, 61; in turgor, 83; penetrating power of, 65, 68, 74, 77. PRESSURE: exudation, 102; gas, 9; os- motic (see Osmotic pressure). PROTOPLASM, 49. PROTOPLASMIC MEMBRANES, action of, 80. PURIEWITCH, 78. QUERCUS, copper in, 69. QUINCKE, 82. RACIBORSKI, 132, 137, 138. REINHARDT, 88. REPRODUCTION, in osmotic solutions, 138. RICHARDS, 68. RICHTER, 139. RISE OF WATER, 107. ROOT HAIRS, exudation from, 75. ROOTS, permeability of, 69. ROTHERT, 140. SALTS, inorganic, penetrating power of, 67, 71. SAMBUCUS, permeability of, 76. SAP : cell, 52 ; expressed, conductivity of, 85. SARGENT, 128. SCHEFFER, 18. SCHNEIDER, 112. SCHWENDENER, 76. SEEDS, in osmotic solutions, 130, 131. SELECTIVE POWER, 118. SIPHONED}, bursting of, 54. SKERTSCHLEY, 69. SLOSSON, 131. SODIUM, penetrating power of, 69. SODIUM CHLORID : in plasmolysis, 61 ; penetrating power of, 68, 74; in me- dium, 137. SODIUM NITRATE: absorption of, 120; penetrating power of, 79. SOLIDS, absorption of, 118 ; diffusion of, 14. SOLANUM, absorption by, 69. SOLUTES: absorption of , 115; active, na- ture of, 83; transmission of, 120. SOLUTION, Knop's, 132. SOLUTIONS: denned, 16; properties of, 124 ; molecular, 21 ; normal, 21 ; of gases in liquids, 17; of liquids in liquids, 16; of solids in liquids, 18; terminology for, 20. SOLUTION THEORY OF SEMIPERMEABIL- ITY, 80. SPERMS, in osmotic solutions, 129, 139. SPIROGYRA : cold plasmolysis of, 75, 141 ; permeability of, 62, 65. 68. STAHL, 140. STANGE, 84, 130. STENTOR, cold plasmolysis of, 75, 143. STEVENS, 69. STICHOCOCCUS, permeability of, 70. STIGEOCLONIUM : drying out of, 143; in osmotic solutions, 132, 138; zoospores in solutions, 139. STOMATA, 93. STRASBURGER, 110. STRATIOTES, permeability of, 62. STRONGYLOCENTUTUS, eggs of, 129. SUCRASE, penetrating power of, 71. SUGAR, CANE : in plasmolysis, 61 ; pene- trating power of, 65. SUNFLOWER, permeabilty of, 65. SUPPORT, mechanical, by turgidity, 87. SURROUNDING MEDIUM, influence of, 124. SUTHERST, 85. TAMMAN, 36. TEMPERATURE: influence of, on absorp- tion, 95 ; influence of, on permeabilty, 75 ; influence of, on turgor, 77. TENSION, diffusion, of solvent, 30. THEORIES : of matter, 3 ; of semipermea- bility, 82. THUNBERGIA, permeability of, 71. THUYA, transpiration of, 111. TONIE, 59. TONOPLAST, 52, 53, 80, 82. TRADESCANTIA, permeability of, 56, 62, 76. TRANSMISSION: of solutes, 115, 120; of water, 95. TRANSPIRATION, 93, 96. TRANSPIRATION STREAM, 107. TRAUBE, 50. TROPISM, reversal of, 139. TRUE, 68, 137. TURGESCENCE, 52. TURGIDITY: nature of, 49, 52, 53; main- tenance of, 86, 121 ; relation of, to activ- ity, 87. TURGOR, 52, 55, 74. TURNIP, pressure of sap of, 85. UREA: in plasmolysis, 62; penetrating power of, 77, 79. VACUOLE, 52. VANDERVELDE, 131. VAN RYSSELBERGHE, 76, 134. VAN'T HOFF, 25, 37, 40. VEGETABLE MARROW, pressure of sap of, 85. VESQUE, 95. VICIA : nectaries of, 79 ; roots of, 65. VIOLET, methyl, penetrating power of, 66. VOIGTLANDER, 18. VON MAYENBURG, 67, 83, 118. WALDEN, 36. INDEX 149 WALKER, 40. WIESNER AND MOLISCH, 116. WALL, cell, 51. 52. WILSON, 98. WATER : absorption of, by plant cells, 91 ; WLADIMIROFF, 58. absorption of, by muscle, 53: in cellu- Woorm 112 lose wall, 52; loss of, 95; transmission " ' Of 95. WORTMANN, 66. WATER PORES, 74, 96. YASUDA, 132. WESTERMEIER, 108. ZEA' solutes of, 83. WHETHAM, 43. ZOOSPORES, in osmotic solutions, 139. WIELER, 65, 102, 108. -P O a CO O -H CQ to g£ IO -P TJ PQ O * 0 Is CQ -P o CQ vaan