OS- 
 
 
 v 
 
 LIBRARY 
 
 . OF THR 
 
 UNIVERSITY OF CALIFORNIA. 
 
 Received.... ...NV-19--1JB91 , 18 
 
 Accessions No. <&.*.&M. _ Shelf No. 
 
 ' Or, 
 
TEXT-BOOKS OF SCIENCE 
 
 ADAPTED FOR THE USE OF 
 
 ARTISANS AND STUDENTS IN PUBLIC AND SCIENCE SCHOOLS 
 
 CHEMICAL PHILOSOPHY 
 
New Edition of Watts' Dictionary of Chemistry. 
 * WATTS' DICTIONARY OF CHEMISTRY. 
 
 Revised and entirely Re-written by H. FORSTRR MORLEY, M.A. 
 D.Sc. Fellow of, and lately Assistant-Professor of Chemistry in, 
 University College, London ; and M. M. PATTISON MUIK, M.A. 
 F.R.S.E. Fellow, and Praelector in Chemistry, of Gonville and Caius 
 College, Cambridge. Assisted by Eminent Contributors. To be 
 published in 4 vols. 8vo. Vols. I. and II. (A Indigo) now ready, 
 price 42J. each. 
 
 THORPE'S DICTIONARY OF APPLIED 
 
 CHEMISTRY. 
 
 By T. E.THORPE, B.Sc. (Viet.) Ph.D. F.R.S. Treas. C.S. Professor 
 of Chemistry in the Normal School of Science and Royal School of 
 Mines, South Kensington. Assisted by Eminent Contributors. 
 3 vols. price 42$. each. Vol. I. (A Dysodil) now ready. 
 
 MODERN THEORIES OF CHEMISTRY. 
 
 By Professor LOTHAR MEYER. Translated, from the Fifth Edition 
 of the German, by P. PHILLIPS BEDSON, D.Sc. (Lond.) B.Sc. (Viet.) 
 F.C.S. Professor of Chemistry, Durham College of Science* and 
 W. CARLETON WILLIAMS, B.Sc. (Viet.) F.C.S. Professor of 
 Chemistry, Firth College, Sheffield. " 
 
 4 This is, perhaps, the most 
 important and profound philo- 
 sophical treatise on one of the 
 most fascinating and widely 
 studied sciences.. .There could 
 
 scarcely be a more acceptable 
 book published than this transla- 
 ation, which commends itself by 
 the manifest care exercised in its 
 production.' SCOTSMAN. 
 
 London: LONGMANS, GREEN, & CO. 
 
INTRODUCTION TO THE STUDY 
 
 OF 
 
 CHEMICAL PHILOSOPHY 
 
 THE PRINCIPLES 
 
 OF 
 THEORETICAL AND SYS TEAT A TIC CHEMISTRY 
 
 BY 
 
 WILLIAM A. TILDE.N, D.Sc.LOND., F.R.S. 
 
 PROFESSOR OF CHEMISTRY IN THE MASON COLLEGE, BIRMINGHAM 
 LATE LECTURER OX CHEMISTRY IN CLIFTON COLLEGE 
 
 EW EDITION 
 
 OF THE 
 
 TJHIVERSIT7 
 
 LONDON 
 
 LONGMANS, GREEN, AND CO. 
 
 AND NEW YORK : 15 EAST i6 Ih STREET 
 1890 
 
 A II rights reserved 
 
PRINTED BY 
 
 SPOTTISWOODE AND CO., NEW-STREET SQUARE 
 LONDON 
 
PREFACE 
 
 TO 
 
 THE SIXTH EDITION. 
 
 TWELVE years have elapsed since the production of this 
 little book. In the course of those years many facts have 
 been added to our knowledge of chemical laws and pheno- 
 mena, and many changes have been introduced into chemical 
 theory. Thus all the gases have been liquefied, several new 
 elements have been discovered, and the * periodic law ' has 
 been accepted by the majority of chemists. Hence it has 
 become necessary to subject the book to very thorough 
 revision. This I have done, but at the same time I have 
 not felt it desirable to alter its character or to increase its 
 dimensions so much as to render it less suitable than before 
 for those persons for whom it was originally designed, 
 namely, for pupils in the advanced classes in schools 
 and colleges, and generally for those students who have 
 passed through a sufficient preparatory course of experi- 
 mental instruction. 
 
 For the student who intends to pursue chemistry 
 specially it may serve as an introduction to more compre- 
 hensive works, such as Professor Lothar Meyer's 'Modern 
 Theories.' 
 
 In matters of theory it can hardly be expected that the 
 
vi Chemical Philosophy. 
 
 views expressed by one writer should agree entirely with 
 those of another. I have, however, done my best to treat 
 in an impartial spirit those parts of the subject upon which 
 opinion is divided. I have also admitted very little matter 
 of a speculative kind, and the single chapter which might be 
 open to such an objection (Chap. XXII.) may be omitted by 
 young students. In a few cases I have felt justified in 
 stepping a little in advance of the ordinary orthodox teach- 
 ing as, for example, in connection with valency. With regard 
 to classification, I have preferred to retain the arrangement 
 of the elements in the most obviously natural groups, as 
 being easier for the student than a system based rigidly 
 upon the periodic law, which, however, is discussed in a 
 separate chapter. 
 
 In the direction of thermo-chemical investigation great 
 progress has been made ; but much yet remains to be done 
 in this department of the science before the full significance 
 of the observed facts will become manifest. Partly on this 
 ground, and partly because the subject could not be 
 treated in an instructive manner without making very 
 extensive additions to the book, I have preferred to omit, 
 for the present, a description of the experimental methods 
 and all but the most obvious conclusions. 
 
 The short sketch of the classification of carbon com- 
 pounds is, of course, wholly inadequate if regarded as 
 representing the range of this extensive department of 
 chemistry. It is introduced merely for the benefit of the 
 general reader who may not have time for the serious study 
 of this complicated subject ; it may also serve as a short 
 syllabus for the use of classes in schools. 
 
 W. A. T. 
 BIRMINGHAM : May 1888. 
 
EXTRACT FROM 
 
 PREFACE TO THE FIRST EDITION. 
 
 THIS little volume is primarily intended for the use of 
 students. It aims at presenting a synopsis brief indeed, and 
 probably imperfect of the leading principles of chemistry 
 in such a form as to give the subject a more decided 
 educational direction than has been hitherto customary. 
 
 In consideration of its peculiar fitness for developing the 
 powers of observation, of reasoning, and of memory, no 
 branch of experimental science deserves more emphatic 
 recognition at the hands of educators than chemistry. In 
 order, however, that its advantages may be reaped to the full, 
 I believe that the methods of teaching very generally preval- 
 ent in schools require to be considerably modified. I think 
 teachers ought to realise the fact that chemistry, as a school 
 subject, is not taught with a view to its practical applica- 
 tions to medicine, manufactures, or the arts, but because the 
 study is calculated to quicken the faculties of observation, 
 to strengthen the memory, and to engender a power and a 
 habit of continuous thought, as well as to arouse new 
 interests and open up new fields to the imagination. It is 
 
viii CJicimcal Philosophy* 
 
 of little consequence, in this view, whether or not the facts 
 acquired can be turned to practical account ; but it is of 
 prime importance that the phenomena brought under their 
 notice, and the manner in which those phenomena are 
 presented, should be such as will compel the pupils to 
 think. 
 
 Notwithstanding that the book does not profess to be a 
 complete treatise on the subject, its contents will, I believe, 
 be found sufficiently comprehensive to afford a tolerably 
 general view of chemical theory as it exists at the present 
 time. My desire has been to assist the student in attaining 
 to broad and philosophic views of chemistry as a whole, 
 and to accustom him to regard it as one out of many 
 branches of physical science rather than as a mystery stand- 
 ing apart from other studies 
 
 CLIFTON : May 1876. 
 
CONTENTS. 
 
 PAG8 
 
 TABLE OF WEIGHTS, MEASURES, AND TEMPERATURES . 2 
 
 SECTION I. 
 CHAPTER I. 
 
 THE CONSTITUTION OF MATTER. 
 
 Molecules Atoms Chemical Affinity Special Province of 
 Chemistry Solids Liquids Gases Intermediate States 
 of Matter .......... 3 
 
 CHAPTER II. 
 FUSION. 
 
 Melting and Dissolution Relation of Melting-point to Molecular 
 Weight Melting of Mixtures Eutectic Mixtures Cryo- 
 hydrates . . . . . . . . .12 
 
 CHAPTER III. 
 
 SOLUTION. 
 
 Solution of Solids Selective Power of Solvents Fusibility and 
 Solubility Relation of Solubility to Temperature Thermal 
 Effects of Dissolution Dissolution generally attended by 
 
Chemical Philosophy. 
 
 PAGE 
 
 Contraction Vapour Pressure of Solutions Solutions of 
 Coloured Salts Theories of Solution Crystallisation of 
 Solutions Water of Crystallisation Supersaturated Solu- 
 tions Solution of Gases Coefficients of Absorption Law 
 of Henry and Dalton Case of Atmospheric Air . . . 16 
 
 CHAPTER IV. 
 LIQUID DIFFUSION AND DIALYSIS. 
 
 Crystalloids and Colloids Dialysis Soluble Colloidal Ferric, 
 
 Silicic, and Aluminic Hydrates, &c 29 
 
 CHAPTER V. 
 
 EVAPORATION AND EBULLITION. 
 
 Conditions of Ebullition Relation of Boiling-point to Mole 
 cular Weight Homologous Series Fractional Distillation 
 Liquefaction of Vapours and Gases Critical Temperature 
 of Gases under Pressure ....... 33 
 
 CHAPTER VI. 
 
 DIFFUSION AND DIALYSIS OF GASES. 
 
 Law of Diffusion Graham's Experiments Explanation of Diffu- 
 sion by Molecular Kinetic Theory Passage of Gases 
 through Colloid Septa Occlusion of Gases by Metals- 
 Absorption of Gases by Porous Solids .... 40 
 
 CHAPTER VII. 
 
 RELATION OF GASES TO TEMPERATURE AND PRESSURE. 
 
 Law of Boyle Law of Gay-Lussac and Charles Examples 
 
 Absolute Temperatures Law of Avogadro ... 46 
 
Contents. xi 
 
 CHAPTER VIII. 
 
 SPECTRA EMISSION SPECTRA. 
 
 PAGE 
 
 Spectra of Solids, Liquids, and Dense Vapours are usually Con- 
 tinuousLine Spectra of Ignited Gas Spectral Analysis 
 Discovery of Elements by Spectroscope Absorption Spec- 
 tra Fraunhofer's Lines Heating and Chemical Effects of 
 Radiant Energy Luminosity of Flame Theories of Davy 
 and Frankland . . 53 
 
 EXERCISES ON SECTION I 67 
 
 SECTION II. 
 
 CHAPTER IX. 
 
 ELEMENTS AND COMPOUNDS. 
 
 The Term Element in Chemistry only Provisional Elemental 
 Character of Chlorine, Potassium, Oxygen, &c., established 
 by Experiments Distribution of Elements in Nature Table 
 of Known Elements, with Symbols and Atomic Weights 
 Formulae and Molecular Weights ..... 70 
 
 CHAPTER X. 
 
 LAWS OF CHEMICAL COMBINATION. 
 
 Definite, Multiple, and Reciprocal Proportions Gay-Lussac's 
 Law of Volumes Composition expressed in Percentages of 
 Elements Dalton's Atomic Theory ..... 79 
 
 CHAPTER XI. 
 
 EQUATIONS CLASSIFICATION OF REACTIONS. 
 
 Calculations of Weight and Volume in Chemical Changes Classi- 
 fication of Reactions Combination of Entire Molecules 
 Decomposition of Molecules Metameric Change Single 
 Displacement Double Decomposition Substitution Com- 
 pounds .......... 86 
 
xii Chemical Philosophy. 
 
 CHAPTER XII. 
 
 CHEMICAL COMPOUNDS DISTINGUISHED FROM MIXTURES. 
 
 PAGE 
 
 Properties of Compounds always Different from those of their 
 Constituents Chemical Compounds have a Definite Compo- 
 sition Fractional Crystallisation of Solids Fractional Dis- 
 tillation of Liquids Fractional Solution and Diffusion of 
 Gases Examples Atmospheric Air is a Mixture . . 92 
 
 CHAPTER XIII. 
 
 NOMENCLATURE. 
 
 Names of Elements Of Binary Compounds Of Acids and Salts 
 
 Of Carbon Compounds . . . . 99 
 
 EXERCISES ON SECTION II. ....... 105 
 
 SECTION III. 
 
 CHAPTER XIV. 
 
 ATOMIC WEIGHTS. 
 
 Experimental Determination of Combining Proportions Oxygen, 
 Carbon, Chlorine, Potassium, Silver, considered as Examples 
 Determination of Atomic Weight by Reference to Density 
 of Gas or Vapour- -Law of Dulong and Petit Law of Iso- 
 morphism Atomic Weights of Oxygen, Carbon, &c., deter- 
 mined by Reference to Constitution of their Compounds . 109 
 
 CHAPTER XV. 
 
 MOLECULAR WEIGHTS AND FORMULAE. 
 
 Formula deduced from Percentage Composition Vapour Density 
 and Molecular Weight Synthetical Formation of Molecules 
 of Oxygen, Hydrogen, &c. Nascent State Molecular For- 
 mulae of Molecules Synthesis and Analysis of Compounds 
 Saturating Power of Acids and Bases Substitution Com- 
 pounds . . . . . . . . . .122 
 
Contents. xin 
 
 CHAPTER xvi. 
 
 TYPES VALENCY. 
 
 PAGE 
 
 Rational and Typical Formulae Constitutional Formulae For- 
 mula of Sulphuric Acid for Example Valency or Combining 
 Capacity- Atomicity Valency of a given Atom^ varies 
 according to Nature of Elements with which it is com- 
 binedValency a Function of Temperature Table of 
 Principal Elements -Examples of Constitutional and 
 Graphic Formulae Law of Even Numbers Artiads and 
 Perissads Compound Radicles Table of Valency of Com- 
 mon Salt Radicles *3 6 
 
 EXERCISES ON SECTION III J 53 
 
 SECTION IV. 
 
 CHAPTER XVII. 
 
 CONDITIONS OF CHEMICAL CHANGE. 
 
 in oraer that Chemical Action must occur between Two Bodies 
 they must be in contact with each other Complete Union 
 only occurs in Liquid or Solid State Influence of Tempera- 
 tureOf the Solvent or Medium Influence of Preponderant 
 Mass Action of a Third Body Catalytic Actions . .161 
 
 CHAPTER XVIII. 
 
 DISSOCIATION. 
 
 Nitrogen Peroxide Ammonium Salts Calcium Carbonate- 
 Efflorescent Salts Dissociation in Solution Abnormal 
 Vapour Densities Explanation of Dissociation Dissocia- 
 tion compared with Evaporation Table of Examples . .168 
 
xiv Chemical Philosophy. 
 
 CHAPTER XIX. 
 
 THERMAL AND ELECTRICAL EFFECTS OF CHEMICAL ACTION. 
 
 FAGS 
 
 Heat of Neutralisation of Acids Effect of Successive Molecules 
 of Base added to Polybasic Acids Chemically Free or Un- 
 saturated Bodies contain a Store of Energy Principle of 
 Maximum Work Heat of Formation of sundry Chlorides, 
 Bromides, Iodides, Oxides Chemical Compounds contain 
 generally Less Energy than the Materials from which they 
 are formed Exceptions Electricity produced by Chemical 
 Action Electrolysis Difficulties presented by some Pheno- 
 mena of Electrolysis . . . . . . . .176 
 
 CHAPTER XX. 
 
 COMBUSTION. 
 
 Modern Explanation of Combustion Theory of Phlogiston 
 Igniting Points Temperature of Flame Heat of Combus- 
 tion Problems 186 
 
 CHAPTER XXI. 
 
 ISOMERISM. 
 
 Physical Isomerides Polymerism Metamerism Theory of 
 Asymmetric Carbon Allotropic Forms of the Elements 
 Energy of Isomerides . . . . . . .192 
 
 CHAPTER XXII. 
 
 THEORIES REGARDING THE NATURE OF CHEMICAL AFFINITY. 
 
 General Characteristics of Chemical ' Affinity' Three Chief Hy- 
 potheses : (i) Chemical Combination the Result of a Special 
 Attraction ; (2) Electro-Chemical Theories ; (3) Atomic 
 Motion Chemical 'Affinity' compared with Adhesion 
 Molecular Compounds Molecular Volume Relative 
 
Contents. xv 
 
 PAGE 
 
 Strength of Affinity between Different Elements The 
 Chemical and Dynamical Equivalent distinguished 
 Avidity of Acids ........ 203 
 
 EXERCISES ON SECTION IV .215 
 
 SECTION y. 
 
 CHAPTER XXIII. 
 
 CLASSIFICATION OF ELEMENTS. 
 
 Division of Elements in Three Classes : Non-metals, Metalloids, 
 Metals Characters of Non-metals Class I. The Halogens 
 Class 2. Oxygen, Sulphur, Selenion Class 3. Boron 
 Class 4. Carbon, Silicon Class 5. Nitrogen, Phos- 
 phorus . . .217 
 
 CHAPTER XXIV. 
 
 DIVISION II. METALLOIDS. 
 
 I. Hydrogen 2. Tellurium 3. Germanium, Tin, Titanium, 
 Zirconium 4. Vanadium, Arsenic, Antimony, Bismuth 
 5. Niobium, Tantalum 6. Molybdenum, Tungsten, Ura- 
 nium .......... 244 
 
 CHAPTER XXV. 
 
 DIVISION III. METALS. 
 
 General Characters of Metals Class I. Metals of the Alkalies 
 Class 2. Metals of the Alkaline Earths Class 3. Zinc 
 Group Class 4. Silver, Mercury Class 5. Thallium, Lead 
 Class 6. Aluminium Group Class "J. Iron-Copper Group 
 Class 8. Platinum Group ...... 256 
 
xvi Chemical Philosophy. 
 
 CHAPTER XXVI. 
 
 THE PERIODIC LAW. 
 
 PAGE 
 
 Trout's Hypothesis Short Series of Related Elements Dumas 
 compares such Series to Homologous Series Newlands 
 Lothar Meyer's Curve, showing Periodicity of Atomic 
 Volume MendelejefPs Classification 284 
 
 EXERCISES ON SECTION V 293 
 
 SECTION VI 
 CHAPTER XXVII. 
 
 CLASSIFICATION OF COMPOUNDS. 
 
 Acids Determination of Basicity of Acids Bases Ammoniacal 
 
 Bases Salts 296 
 
 CHAPTER XXVIII. 
 
 DERIVATIVES OF AMMONIA. 
 
 Ammonium Theory Amines Constitution of Ammonium Com- 
 pounds Phosphines, Arsines, Stibines Amides . . 307 
 
 CHAPTER XXIX. 
 
 CARBON COMPOUNDS. 
 
 Homologous Series Hydrocarbons and their Haloid Derivatives 
 Alcohols Primary, Secondary, Tertiary Alcohols Phenols 
 Ethers Aldehyds and Ketones Acids Basic Deriva- 
 tives of Ammonia Compound Ethers, or Ethereal Salts 
 Organo-metallic Compounds . . . . . . 314 
 
 EXERCISES ON SECTION VI 334 
 
 INDEX. . 339 
 
CHEMICAL PHILOSOPHY. 
 
METRIC WEIGHTS AND MEASURES. 
 
 PREFIXES. Myria- 10,000 
 
 Kilo- 1,000 
 
 Hecto- 100 
 
 Deka- 10 
 
 Deci- ..'.'...'..' fJ 5 
 Centi- ........ yjj 
 
 Milli- T J i)ff 
 
 Inserting after the hyphen and in the brackets 
 
 (i.) Metre, gives the measures of length. 
 
 (2.) Litre, capacity. 
 
 (3.) Gram, ,, weight. 
 
 LENGTH. 
 
 i METRE = 10 decimetres. (10 dcm.) 
 
 = 100 centimetres. (100 cm.) 
 
 = 1,000 millimetres. (1,000 mm.) 
 1,000 Metres = i Kilometre. 
 
 CAPACITY. 
 
 i LITRE = I cubic decimetre. (i c.dcm.) 
 
 = 1,000 cubic centimetres. (1,000 c.c.) 
 
 WEIGHT. 
 
 i GRAM or the weight of i c.c. of pure water at 4 C. 
 
 = 10 decigrams. (10 dg.) 
 
 = 100 centigrams. (100 eg.) 
 
 = 1,000 milligrams. (1,000 nig.) 
 ijOoo Grams = i Kilogram. 
 
 ENGLISH EQUIVALENTS. 
 
 NEARLY ACCURATELY 
 
 I METRE = 3 feet 38 inches = 39*37079 inch' s. 
 
 or 39$ inches. 
 
 i KILOMETRE = 1,100 yards = 1093*633! yards. 
 
 i LITRE = if Imp!, pint = i*76 0773414 pint. 
 
 or 35 ounces. 
 
 I GRAM =15^ grains = i5*43 23488 grains, 
 
 i KILOGRAM = s pounds = 2*2o 402 i 3 pounds. 
 
 MEMORANDA. 
 
 Weight of i litre of hydrogen (at o C. and under 760 mm. bar. 
 pressure) = '0896 gram, or i crith. 
 
 Therefore i gram of hydrogen measures 11*16 litres. 
 
 Specific gravity of Hydrogen (Air=i) = '0693. 
 Air (H = i) = 14*42. 
 
 5 C. = 9 F. 
 .'. i C. = f F., and i F. = 
 C x 
 
 To convert C. into F. temperatures 
 
 /r* 
 
 To convert F. into C. temperatures 
 
SECTION I. 
 
 CHAPTER I. 
 
 THE CONSTITUTION OF MATTER. 
 
 IN order to facilitate the explanation of chemical pheno- 
 mena, modern philosophers have found it convenient to 
 revive, in a somewhat modified form, the ancient hypothesis 
 that all bodies possessing extension and weight are made 
 up of stuff, substance, or matter, which is not uniform and 
 continuous throughout, but consists of separate very small 
 portions. Each of these small masses, which are called 
 molecules, is supposed to be to a certain extent independent 
 of the rest and isolated from them. The hypothesis further 
 requires us to suppose that the molecules constituting any 
 given species of matter are all alike, in size, weight, and 
 properties, and differ in these respects from other molecules. 
 Thus the molecules contained in one drop of water are 
 conceived to be precisely like the molecules in any other 
 drop of the same liquid. Similarly the molecules in a given 
 globule of the liquid metal mercury, or quicksilver, must be 
 assumed to be like all other molecules of the same metal ; 
 but water molecules and mercury molecules differ altogether 
 from each other in weight and chemical properties. It 
 must be distinctly understood that molecules cannot be 
 seen, and all the arguments upon which the assumption of 
 their existence is founded are derived from the examination 
 of masses of appreciable magnitude. We know nothing of 
 isolated individual molecules ; the examination of these 
 
 B 2 
 
4 Chemical Philosophy. 
 
 would be for obvious reasons impossible, and even if such a 
 division were actually possible, the condition or properties 
 of a single molecule would be in no way comparable with 
 those of a mass of matter in which many molecules are 
 imagined to be naturally aggregated by cohesion or other- 
 wise. But we may assume that when one kind of matter 
 affects chemically another kind, the smallest quantity of 
 each which is capable of entering into the reaction consists 
 of a determinate number of molecules. For the purposes 
 contemplated in this book, then, a molecule may be defined 
 as the unit of chemical action ; that is, the smallest quan- 
 tity which is able to take part in or result from a chemical 
 change. 
 
 When two different kinds of matter are mixed together, 
 chemical action often takes place between them. This 
 action is supposed to be due to the union of molecules of 
 the one kind with molecules of the other kind, or to an 
 interchange of their components. The constituent parts of 
 molecules 1 are called atoms? Nothing is known concerning 
 the true nature of atoms, though they have at different 
 times been supposed to be hard particles of various geome- 
 trical forms, or vortices, like rings in air often produced from 
 the top of a chimney, or from the funnel of a locomotive, 
 and rendered visible by the smoke or steam. To the 
 chemist the word simply means a small mass which cannot 
 be divided, or, at least, has never been known to be divided, 
 into smaller masses in the course of chemical action. An 
 atom, then, is the smallest portion of matter which can be 
 transferred from one molecule to another. 
 
 When the atoms within a molecule are all alike, the 
 substance made up of such molecules is called an element. 
 Most commonly, however, there are reasons for considering 
 that the atoms composing a given molecule are dissimilar, 
 and then the body is a compound. The number of atoms 
 
 1 Molecule, diminutive, from Lat. moles, a mass. 
 
 2 Atom, from a, not, and Te>y, I cut. 
 
Atoms and Molecules. 5 
 
 in a molecule is very variable. A few elements have mole- 
 cules which are assumed to consist of one atom only, others 
 of two, three, or four ; whilst in compound bodies the 
 number may amount to hundreds. 
 
 Atoms are held together within the molecule, and pro- 
 bably molecules also are in certain cases united together by 
 the operation of what is called, for want of a better term, 
 chemical affinity, or chemical attraction. Of the nature of 
 chemical attraction nothing is known. It cannot operate 
 at a distance like gravitation, but requires that bodies should 
 be in contact before they can combine together or act upon 
 each other. It is apparently opposed by heat, for a suffi- 
 cient elevation of temperature serves to break up the mole- 
 cules of all compounds, and even those of many elementary 
 substances, and in this disruption of molecular or atomic 
 combinations by heat there is a very close analogy with the 
 process of evaporation of a liquid. (See Dissociation.) It 
 is closely related to electricity, and an electro-chemical 
 theory, proposed by Davy in 1806, though never fully 
 accepted, contains an element of probability which has led 
 to its revival in a modified form in more recent times. 
 This theory attributes chemical combination to the existence 
 of charges of electricity of opposite kinds upon the atoms 
 which unite, but there are several difficulties in the way 
 which have never been satisfactorily cleared up, and are 
 still a subject of debate. This question will be referred to 
 again in a later chapter. 
 
 The study of those forces which affect entire molecules, 
 and masses of molecules without regard to their composi- 
 tion, belongs to the domain of Physics. The immediate 
 object of Chemistry is to ascertain the composition of 
 bodies by the application of methods of analysis or of syn- 
 thesis, but the science is not confined to the practice of 
 these experimental arts. It has long been known that the 
 same elements united in the same proportions do not neces- 
 sarily produce compounds having in every case the same 
 
6 CJieuiical Philosophy. 
 
 properties in other words, the order in which atoms are 
 united together is not less important than their number, 
 and constitution must be clearly distinguished from composi- 
 tion. (See Isomerism.) 
 
 The density, fusibility, volatility, colour, crystalline form, 
 and other physical properties of both elementary and com- 
 pound bodies are dependent upon the nature, number, and 
 order of the atoms composing their molecules, and it seems 
 to be now fully established that the chemical characters of 
 an elementary atom are directly related to its mass, as 
 represented by the value of its atomic weight. Further- 
 more, the combination and the separation of atoms or of 
 molecules is attended by a distribution of energy, and every 
 chemical change is attended by absorption or evolution of 
 heat, and frequently by other physical phenomena. 
 
 Chemistry is therefore a science of observation and 
 of experiment, seeking by the study of the properties of 
 bodies to establish a system of classification, and to connect 
 that system with a knowledge of their constitution, both in 
 regard to matter and the quantity of energy associated with 
 it. When the science has made some further advances it 
 may become possible more perfectly than at present to 
 infer the constitution of a body from a knowledge of its 
 properties, and on the other hand to predict with greater 
 certainty the properties of the compound that would result 
 from the union of given materials in any predetermined 
 order. 
 
 The materials which compose the earth's crust, with its 
 ocean and atmosphere and their inhabitants, may be roughly 
 classified according to their mechanical condition into solids 
 and fluids. Fluids are either liquids or gases. 
 
 A solid retains its form unless acted upon by pressure, 
 by division with cutting instruments, by heat, or by solvents. 
 Many solid bodies under suitable conditions assume definite 
 geometric figures, which are generally bounded by plane 
 faces, and thus give rise to crystals. This rigidity, by which 
 
Solids and Fluids. 7 
 
 the external form and relative position of parts is maintained, 
 is due to what is called 'cohesion.' We should not, how- 
 ever, be justified in assuming that the constituent particles 
 of solids are in a condition of absolute repose. Alteration 
 of temperature and volume produced by the application of 
 heat are attributable, according to the molecular theory, to 
 the motion of the molecules, this motion being supposed to 
 consist mainly in their rotation, or oscillation, or revolution, 
 within a determinate space. 
 
 A liquid is recognised by its mobility, and by always 
 assuming when at rest a horizontal level surface, except just 
 where it comes into contact with the vessel containing it. 
 The several parts of a mass of liquid are not held together 
 with the same amount of force that binds together the 
 parts of a solid, for it is generally much less difficult to 
 detach a drop from a mass of liquid than to break off a 
 portion of a solid. The curved surface of a liquid at con- 
 tact with a solid, the ascent of liquids in capillary tubes, and 
 the spheroidal form of the rain-drop as well as of water 
 sprinkled upon a greasy surface, are all effects of * surface 
 tension ' that is, these phenomena are due to the fact that 
 the superficial layers of a liquid behave like an elastic 
 membrane stretched over the liquid and exerting a pressure 
 upon it. The volume of a liquid is scarcely affected 
 appreciably by pressure, even when very great, and pressure 
 communicated to any one part of a mass of liquid is trans- 
 mitted almost instantly and without loss to every other part. 
 
 From the phenomena of liquid diffusion which will be 
 alluded to further on, it seems probable that the molecules 
 of liquids are constantly moving about. These motions, 
 however, are sluggish in comparison with the corresponding 
 intestine movement which is observed in gases, and this is 
 perhaps to be explained by the assumption that the mole- 
 cules of liquids are comparatively close together, so that 
 their free motion is impeded by frequent collision. 
 
 Gases differ from solids and liquids in the circumstance 
 
8 Chemical Philosophy. 
 
 that they seem to be entirely discharged from the influence 
 of cohesion. A mass of gas exhibits no surface, like that of 
 a liquid, and no gas can be confined in a vessel which does 
 not enclose it on every side. 
 
 The volume of a gas increases as the pressure upon it 
 decreases (Law of Boyle), until when the pressure is nothing 
 the bulk of the gas becomes greater than the capacity of any 
 conceivable vessel. If, therefore, a small quantity of a gas 
 be introduced into any part of a vacuous space, it imme- 
 diately spreads itself out and pervades every portion of that 
 space equally. 
 
 Solids, liquids, and gases alike expand upon the applica- 
 tion of heat, but whereas each solid and liquid increases by 
 a fraction of its volume which is peculiar to itself, all gases 
 expand to practically the same extent by equal increment of 
 temperature. In other words, whilst the coefficients of 
 expansion of solids and liquids are all different, those of 
 true gases are the same in every case. 
 
 On the hypothesis that a gas, like a solid or a liquid, is a 
 congeries of small masses or molecules, there are reasons 
 for supposing that the molecules of gases are in constant 
 and very rapid motion from place to place, and that they 
 move in straight lines. A given molecule, however, cannot 
 be supposed to pursue an uninterrupted course for any 
 appreciable distance, but probably takes a new direction on 
 the approach of another molecule. 
 
 These motions of translation, which, in accordance with 
 received views, have been here attributed to the molecules 
 of gases and liquids, are directly related to the amount of 
 heat which has been expended in producing the liquid from 
 a solid, or the gas from a liquid. In other words, when 
 heat is consumed in melting a solid, or in gasifying a liquid, 
 it is necessary to admit that the vibratory motion to which 
 the temperature of the heating agent is supposed to be due, 
 being communicated to the molecules of the object, may 
 cause them not only to vibrate, but to move from place to 
 
The Molecular Theory. 9 
 
 place ; and this altered motion corresponds with change of 
 state. 
 
 For the further discussion of these speculations the reader 
 must consult works on physics. 1 He will do well, however, 
 constantly to set before his mind the fact that we possess 
 at present no direct or positive proof even that molecules 
 exist ; still less have we any evidence regarding the condi- 
 tions under which they may subsist in mass. The molecular 
 hypothesis, however, is no longer in the position which it 
 formerly held as a relic of the vague speculative philosophy 
 of the ancients. It has been raised to the rank of a theory 
 which bids fair to rival in completeness and importance the 
 Newtonian theory of gravitation itself. In neither case does 
 the theory admit of direct experimental proof; but both are 
 accepted because they accord fully with the results of obser- 
 vation. The theory of molecules once admitted, all the 
 recognised laws of chemical combination by weight and 
 volume follow as necessary consequences. At the same 
 time the phenomena connected with the physical properties 
 of gases and liquids, such as the transmission of pressure 
 and the remarkable laws of diffusion, find a rational and 
 intelligible explanation such as no other hypothesis yet put 
 forward 'has been competent to furnish. It is, in fact, not 
 too much to assert that the rapid progress of modern 
 chemistry and the intimate connection which it has been 
 shown to have with other branches of physical science, as 
 well as the illustrations it affords of the great doctrine of 
 energy, are largely attributable to the general acceptance 
 of this hypothesis. The science as it now stands may be 
 regarded as a practical development of the molecular theory. 
 
 1 The mathematical investigations of Kronig and Clausius, and of 
 Rankine and Clerk Maxwell, have led to the establishment of the 
 kinetic theory of gases, and have given a powerful impetus to the 
 general recognition of the molecular theory by physicists. The student 
 is recommended to make himself acquainted with the discussion of the 
 subject from this point of view, which is given in Clerk Maxwell's 
 Theory of Heat. 
 
io Chemical Philosophy. 
 
 When a solid is transformed into a liquid, or a liquid 
 into a gas, an apparently abrupt change of physical proper- 
 ties occurs, and heat is abundantly absorbed without pro- 
 ducing elevation of temperature. But the transition from 
 one state to another is by no means so sudden as appears 
 from the consideration of cases like that of water. Many 
 solids, such as iron, pass through an intermediate state, in 
 which they are more or less plastic or viscid before they 
 finally assume the liquid condition ; and even the most 
 perfect liquids with which we are acquainted are far from 
 being absolutely mobile. Ether and alcohol, for example, 
 flow more easily than water ; but even these liquids exhibit 
 a certain degree of viscosity. 
 
 Experiments commenced in 1822 by Caignard de la 
 Tour, and since continued and extended by Dr. Andrews, 
 have shown that matter is capable of existing in a somewhat 
 analogous condition intermediate between the liquid and 
 gaseous states : ' By partially liquefying carbonic acid gas 
 by pressure and then raising the temperature to 88 R, the 
 surface of demarcation between the liquid and gas becomes 
 fainter, loses its curvature, and at last disappears. The 
 space is then occupied by a homogeneous fluid, which 
 exhibits, when the pressure is suddenly diminished or the 
 temperature slightly lowered, a peculiar appearance of 
 moving or flickering striae throughout its entire mass. At 
 temperatures above 88 F., no apparent liquefaction or 
 separation into two distinct forms of matter could be 
 effected, even when a pressure of three or four hundred 
 atmospheres was applied.' (Andrews.) Nitrous oxide and 
 sulphurous oxide, and other gases, give similar results. The 
 striae referred to are most probably the result of changes in 
 density, caused by slight changes of temperature or pressure, 
 as in ordinary liquids or gases when heated. 
 
 It thus appears that the various physical states of matter 
 merge one into another by imperceptible gradations ; and 
 if we adopt the molecular theory we can see some explana- 
 
The Molecular Theory. \ i 
 
 tion of this. The change of a solid into a liquid, and of a 
 liquid into a gas, is the result of alteration, generally increase, 
 in the distances between the molecules and in the rapidity 
 and character of their movements. The generally received 
 opinion is something like the following. In a solid every 
 molecule is attached to the surrounding molecules by cohe- 
 sion, and it is free to move only about a certain mean 
 position, which it never leaves so long as the body remains 
 solid. In a liquid each molecule clings to its neighbours, 
 but less firmly than in the solid, and it is free to move and 
 does move about in the mass, though proceeding in any one 
 direction a very small distance before its path is altered by 
 encounter with other molecules. In a gas the molecules 
 are independent of each other, and move with various 
 degrees of velocity in straight lines. When two molecules 
 approach each other the path of each molecule is altered. 
 In the intermediate states of matter there is probably a 
 mixture of these conditions. Thus, in a liquid at its * critical 
 point,' like carbonic acid gas at 88 F., some molecules are 
 probably moving singly and independently, like the mole- 
 cules of a gas, in the midst of masses of others which are 
 closer together and moving more slowly, after the manner of 
 molecules in a liquid, the relative proportion of molecules 
 in these two conditions serving to determine whether the 
 mass exhibits the character of a gas or a liquid. 
 
 Experiments by Mr. Crookes on the phenomena exhi- 
 bited by highly attenuated gases have led -to the interesting 
 discovery that when the tension of a gas is much reduced it 
 possesses properties wholly different from those of the same 
 gas under ordinary pressures. Under these circumstances, 
 the number of molecules present in a given space being 
 greatly diminished, the distance through which any one of 
 them can move without collision is proportionately increased. 
 When the pressure is reduced to about one-millionth of an 
 atmosphere, and the discharge from an induction-coil is 
 passed through the tube containing the gaseous residue, no 
 
12 Chemical Philosophy. 
 
 luminosity is observed in the contents of the tube, as occurs 
 when the electric discharge is passed through an ordinary 
 vacuum, but the molecules driven from the negative pole 
 with great velocity traverse the whole space before them in 
 straight lines, and striking upon the surface of the glass or 
 upon other solid bodies placed in their path, cause them to 
 emit a phosphorescent light. The stream of molecules 
 flowing thus from the negative pole is also capable of moving 
 light bodies suspended in its course, and when brought to a 
 focus and allowed to impinge upon the surface of a solid it 
 heats it strongly. These phenomena are exhibited equally 
 well by hydrogen, carbonic acid, or atmospheric air, and 
 appear to have no connection with the chemical composition 
 of the gas. It appears therefore that ordinary matter is 
 capable of subsisting in a fourth state, differing as much 
 from the solid, liquid, and gaseous states as these do from 
 one another. 
 
 CHAPTER II. 
 
 FUSION. 
 
 WHEN, by the application of heat, a dry solid, such as 
 sulphur or lead, is made to assume the liquid state, it is 
 said to melt or undergo fusion. But when sugar is placed 
 in water, it disappears, and is said to dissolve, and the liquid 
 which results from such combination is called a solution. 
 This distinction in terms is necessary, if only for practical 
 convenience. 
 
 Every substance melts at a definite temperature, which 
 is always the same for the same substance if pressure is 
 constant. The slight changes of atmospheric pressure have no 
 appreciable effect, but the influence of greater pressures may 
 be recognised by the thermometer. Substances which, like 
 ice, contract on liquefying, have their melting points lowered 
 by pressure, whilst wax and sulphur, which expand on lique- 
 fying, have their melting points raised by pressure. 
 
Fusion of Solids. 13 
 
 When fusion commences the application of heat causes 
 no rise of temperature till the whole is melted. 
 
 It may be added that many substances, after being 
 melted, require to be cooled below the melting point before 
 they will solidify again, unless a particle of the solid is 
 dropped into the liquid, when solidification at once com- 
 mences and the temperature rises to the melting point. 
 Even water, if kept quite still, may be cooled several de- 
 grees below zero without freezing. On shaking or dropping 
 in a particle of ice it begins to crystallise and the tem- 
 perature rises to o, and remains there till the whole is 
 frozen. 
 
 Amongst the elements the temperatures at which fusion 
 occurs are very diverse, and there are but few cases in 
 which any relation can be traced between this property and 
 the chemical or other characteristics of the body. But 
 among the compounds of carbon, which are very numerous, 
 the determination of the melting point often serves as a 
 convenient test whereby to distinguish two similar bodies 
 from one another or to complete the identification of some 
 substance under examination. As a general rule, it may, 
 perhaps, be said that in a series of similar bodies those of 
 smallest molecular weight melt at the lowest temperatures, 
 as in the following examples : 
 
 Melting Point Molecular Weight 
 
 Sulphur 115 n 32 
 
 Selenion 217 n79'5 
 
 Tellurium about 500 n i26 
 
 Formic acid o 46 
 
 Acetic acid 17 60 
 
 Palmitic acid 62 256 
 
 Stearic acid 6g'2 284 
 
 Cerotic acid 78 410 
 
 Melissic acid 88 452 
 
f4 Chemical Philosophy. 
 
 There are, however, numerous exceptions to this rule. 
 Thus in the two following cases the melting points are in the 
 inverse order of the molecular weights : 
 
 Melting Point Molecular Weight 
 
 Cadmium 3 r 5 * 12 
 
 Zinc 423 65 
 
 Magnesium low red heat 24 
 
 Potassium iodide 634 166 
 
 Potassium bromide 699 119 
 
 Potassium chloride 734 76*5 
 
 In other series there is no direct relation between the 
 molecular weight and the melting point of the substance. 
 One example will suffice : 
 
 Melting Point Molecular Weight 
 
 Barium nitrate 593 261 
 
 Strontium nitrate 645 2II '5 
 
 Calcium nitrate 561 164 
 
 In very many cases a mixture of two or more substances 
 melts at a lower temperature than either of the ingredients. 
 
 Mixtures of the fatty and other acids melt at lower 
 temperatures than the pure acids ; the carbonates of potas- 
 sium and sodium melt more easily when mixed than when 
 alone ; an alloy of potassium and sodium is liquid at the 
 ordinary temperature ; and an alloy of cadmium, tin, lead, 
 and bismuth melts in hot water. 
 
 Further, it has been found that when two fusible sub- 
 stances are mixed together in certain proportions the melt- 
 ing point reaches a minimum that is, the mixture melts at 
 a temperature lower than the melting point observed when 
 any other proportions are used. The following diagram 
 represents the melting points of lead and tin and various 
 mixtures of the same. 
 
 It will be noticed that whilst the melting point of tin is 
 228, the temperature of liquefaction is lowered to 187 by 
 mixing four atoms of tin with one atom of lead, although 
 
Eutectic Mixtures. 15 
 
 the melting point of the latter is considerably above that of 
 tin namely, at 326. And upon trial it is found that if lead 
 and tin are mixed together in the proportions of very nearly 
 2 parts of tin to i part of lead, or 3-3 atoms of tin to i atom 
 of lead, the melting point is reduced to 180, and no alloy 
 containing the same metals in any other proportion melts at a 
 
 lower temperature. This sort of mixture of minimum melting 
 point has been called a eutectic mixture, and the pheno- 
 menon has been called * eutexia.' 
 
 There appears to be little doubt that the lowering of the 
 freezing point of water by the addition of soluble salts is a 
 phenomenon of the same order. It is well known, for 
 example, that sea water does not freeze at the same tem- 
 perature as fresh water, and a strong brine remains liquid at 
 
1 6 Chemical Philosophy. 
 
 20 below freezing point. Guthrie found that salts gene- 
 rally are capable of forming eutectic mixtures with water. 
 Thus, if a solution of ammonium chloride, saturated at the 
 ordinary temperature of the air, is cooled down, crystals of 
 the anhydrous salt separate out. If these are separated 
 from the solution as they are formed, the liquid on reaching 
 a temperature of 15 solidifies into a crystalline mass, 
 which consists of i molecule of ammonium chloride united 
 with 12 molecules of water. From this compound neither 
 salt nor ice can be separated, and it melts and solidifies like 
 a homogeneous mass. A compound of this kind was called 
 by Guthrie a ' cryohydrate.' 
 
 The following are some further examples of these com- 
 pounds : 
 
 Temperature 
 Cryohydrate of Solidification 
 
 ZnSO 4 + 2oH 2 O . . . . 7 
 
 . . . . -6 
 
 .... -27 
 
 . . . . -2 
 
 + 22 2 H 2 .... -0-5 
 
 CHAPTER III. 
 
 SOLUTION. 
 
 Solution of Solids. In discussing the questions relating 
 to the phenomena of solution the following facts, among 
 others, must be taken into account. 
 
 i. Considering any solid chosen as an example, we find 
 that it is not soluble in all liquids. For example, common 
 salt is readily and abundantly soluble in water, to a smaller 
 extent in alcohol, and not at all in oils. On the other hand 
 no liquid is capable of taking up every kind of solid. Water 
 
Solution. 17 
 
 dissolves sugar, common salt, nitre, and saline substances 
 generally, but it has no appreciable solvent action upon 
 many common metals and minerals, such as iron or copper, 
 limestone or quartz, neither does it dissolve fat, oil, nor 
 hydrocarbons generally. 
 
 Again, alcohol dissolves many carbonaceous substances, 
 such as resins, phenols, acids, which are almost unaffected 
 by water. Ether and benzene dissolve freely fats and solid 
 hydrocarbons. Carbon bisulphide, which dissolves no salts, 
 takes up many carbonaceous substances, and is the best 
 solvent for common sulphur. Phosphorus trichloride and 
 tribromide dissolve phosphorus. Fluid mercury dissolves 
 many metals, but no other kind of substance. 
 
 There seems to be little doubt, in fact, that solubility 
 is often dependent in some degree upon the existence of a 
 similarity in composition between a solvent and the solid 
 it dissolves. Similia similibus solvuntur. This is further 
 illustrated as follows: 
 
 Salts which contain water of crystallisation are, with 
 comparatively few exceptions, easily soluble in water. Such 
 compounds may be regarded as closely resembling water 
 itself. A molecule such as MgSO 4 7H 2 O may, in fact, be 
 considered as a congeries of eight water molecules, in which 
 one molecule of water is replaced by the elements of the 
 salt. Again, the lower terms of the various series of alcohols 
 and carbon acids show considerable similarity to water in 
 their general behaviour, the higher terms much less. The 
 lower terms of a series like the ethylic series of alcohols 
 may be regarded as consisting of the elements of water, 
 having one atom of hydrogen replaced by a hydro-carbon 
 group, thus 
 
 "^ \ O 
 H/ L 
 
 25 o 3 
 
 H H L H 
 
 These are miscible with water in all proportions. But when 
 the hydrocarbon radicle becomes larger and more compli- 
 
 c 
 
1 8 Chemical Philosophy. 
 
 cated, the resemblance to water and the miscibility with water 
 is less. Thus 
 
 H H / H 
 
 are compounds of the same series as the foregoing, but they 
 are not capable of mixing with water save in limited quan- 
 tity. On reaching the highest known terms of the same 
 series, we find waxy solids which are insoluble in water, but 
 dissolve easily in ether or in hydrocarbon liquids, and them- 
 selves bear a strong resemblance to solid paraffin. The 
 alcohol, Ci 6 H 33 HO, may be regarded as formed of the 
 elements of the hydrocarbon, C 16 H 34 , in which H is 
 replaced by HO, the latter, however, forming a very small 
 proportion of the entire molecule. In the following series 
 of compounds it will be noticed that the solubility in water 
 increases as the amount 'of water residue, HO, present in the 
 molecule increases. 
 
 Name and Formula. Solubility in Water. 
 
 Benzene C 6 H 6 . . . insoluble. 
 
 Phenol C 6 H 5 (HO) . . slightly soluble. 
 
 Catechol ") 
 
 Quinol VC 6 H 4 (HO) . , easily soluble. 
 Resorcinol J 
 
 Pyrogallol C fi H 3 (HO) 3 . . still more soluble. 
 
 2. The fusibility of a substance has much to do with its 
 solubility. Thus, in comparing together three such very 
 similar salts as chloride, bromide, and iodide of potassium, 
 it is found that at all observed temperatures the most fusi- 
 ble, namely, the iodide, m.p. 634, is more soluble in water 
 than the bromide, m.p. 699, and this again is more soluble 
 than the chloride, m.p. 734. However, it by no means 
 follows that fusibility alone confers the property of solubility, 
 for while sodium chloride dissolves abundantly in water, 
 silver chloride, though much more easily fusible, is insoluble. 
 The effect of fusibility is best observed when pairs of sub- 
 
Solution. 
 
 stances which are both soluble to some extent in a given 
 menstruum are compared together at various temperatures. 
 3. This leads to another relation very generally observed, 
 namely, that the solubility of solids usually increases with 
 rise of temperature; in other words, a hot liquid dissolves a 
 given solid more freely than the same when cold. Thus 
 100 parts of water dissolve 
 
 Of KNO 3 KC1 KC1O 3 NaCl 
 
 At o 13-3 29-2 3-3 357 
 
 At 100 247*0 56*5 56*5 39'8 
 
 These relations are best exhibited graphically in the 
 following manner. 
 
 Here the divisions on the horizontal line represent tem- 
 peratures, whilst those on the vertical express parts of salt 
 dissolved at the several temperatures in 100 parts of water. 
 
 C 2 
 
2O Chemical Philosophy. 
 
 It will be noticed that the solubility of the easily fusible 
 chlorate (m,p. 359) increases rapidly with rise of tempera- 
 ture, whilst on the other hand the potassium chloride (m.p. 
 734) and sodium chloride (m.p. 772), which melt at 
 higher temperatures, exhibit only a slight increase of solu- 
 bility under the same circumstances. 
 
 To the rule that solubility increases with rise of tempera- 
 ture there are many apparent exceptions. Thus lime, cal- 
 cium sulphate, and sodium sulphate are partly precipitated 
 when a solution of each, saturated in the cold, is heated to 
 boiling. The case of sodium sulphate is given in the dia- 
 gram. It deserves to be noted, however, that the crystals 
 which fall when a saturated solution of sodium sulphate is 
 heated are anhydrous, whereas the crystals deposited at 
 temperatures below 33 contain ioH 2 O. 
 
 4. When a solid dissolves in a liquid there is always a 
 rise or a fall of temperature in the mass. Moreover, when 
 the solution thus formed is mixed with more water there is 
 again evolution or absorption of heat. The fall of tempera- 
 ture observed when such solids as common salt or nitre are 
 dissolved in water is probably owing to the change from the 
 solid to the liquid state. Many freezing mixtures, as of salt 
 and snow, or sodium sulphate and strong hydrochloric acid, 
 act by liquefying. On the other hand, when dry calcium 
 chloride or sodium sulphate, or sodium carbonate dissolves 
 in water, heat is produced, some of which, at least, is due 
 to the chemical combination of the salt with a portion of the 
 water. 
 
 In the following diagram is represented the thermal 
 effect of adding successive doses of water in molecular pro- 
 portions to sulphuric anhydride. The calorie is an amount 
 of heat sufficient to raise the temperature of one part of 
 water one degree. The divisions of the horizontal lines 
 represent the number of water molecules added, and those 
 on the vertical show the amount of heat evolved at any one 
 stage, or in the entire process. 
 
Solution. 
 
 21 
 
 It will be seen that the thermal change becomes less and 
 less as the liquid is diluted, but even when several hundred 
 molecules of water have been added, the addition of one 
 more produces an effect which can be recognised by the 
 thermometer. 
 
 5. In the act of solution of solids, and especially of an- 
 hydrous salts in water, the volume of the solution is always 
 less than the sum of the volumes of the solid and its solvent, 
 with the exception of some ammonium salts, in which case 
 it is greater. Similarly, the addition of water to a solution 
 is followed by contraction. 
 
22 
 
 Chemical Philosophy. 
 
 The diagram already given exhibits the amount of con- 
 traction observed in the case of sulphuric acid and water, 
 and the following table shows the observed and calculated 
 molecular volumes by which the amount of contraction is 
 indicated. 
 
 
 Molecular Volumes of SO 3 +H a O 
 
 Contraction = Differ- 
 
 
 
 Density 
 
 Mol. Wt. 
 Mol - Vol - = DeSsUy 
 
 ence between calcu- 
 lated and observed 
 Mol. Vol. 
 
 o 
 
 I -940 (Solid. ) 
 
 41-23 
 
 
 
 I 
 
 1-842 
 
 53-20 
 
 6-03 
 
 2 
 
 1774 65-39 
 
 11-84 
 
 3 
 
 I-652 
 
 81-11 
 
 14-12 
 
 4 
 
 1-547 
 
 98-25 
 
 14-98 
 
 5 
 
 1-475 
 
 115-25 
 
 I5-98 
 
 10 
 
 1-286 
 
 202-18 
 
 19-05 
 
 The contraction consequent on the first addition is 
 greater than the second, and for each succeeding molecule 
 is a diminishing quantity, becoming rather less than i after 
 addition of 3H 2 O. 
 
 A similar result ensues if we calculate the contraction 
 following upon the dilution of liquid H 2 SO 4 with successive 
 quantities of water. 
 
 The true explanation of this contraction is unknown, as 
 also is the cause of the change of temperature always ob- 
 served. But in both cases the effect is produced, not per 
 saltum, but continuously, though in gradually decreasing 
 amount as the water is added. 
 
 6. When a solid is dissolved in a volatile liquid the 
 evaporation of the liquid is impeded, the pressure of the 
 escaping vapour is reduced, and the boiling-point of the 
 liquid is raised. 
 
 The table given on next page shows the effect of adding 
 common salt to water. 
 
 A saturated solution of common salt under the pressure 
 of one atmosphere boils at 108-4. 
 
Solution. 
 
 Pressures of vapour evolved from pure water, and reduction of pressure on 
 adding n molecules of NaCl to 100 HjO at temperatures i 
 
 e= 
 
 70 
 
 75 
 
 80 
 
 85 
 
 90 
 
 95 
 
 Water 
 
 = 2 
 
 = 4 
 n = S 
 
 mm. 
 233'3 
 
 18-0 
 
 22-6 
 
 mm. 
 288-8 
 
 10-8 
 
 22'5 
 28-5 
 
 mm. 
 354-9 
 137 
 28-0 
 
 35'2 
 
 mm. 
 
 433-2 
 16-4 
 33-8 
 43-o 
 
 mm. 
 525-5 
 20'4 
 40-9 
 52-5 
 
 mm. 
 6337 
 24-6 
 50-2 
 63-3 
 
 7. Coloured substances usually produce solutions of the 
 same or very similar colour. But metallic salts which are 
 capable of forming crystallised compounds with water often 
 exhibit a variety of colours, according to the amount of 
 water in which they are dissolved. Thus, dry copper sul- 
 phate is white, but it forms a blue solution in water. Cupric 
 bromide is a black substance resembling iodine ; a strong 
 solution of it in water is dark brown, a weak solution pale 
 blue. Cobalt chloride is blue, but when dissolved in water 
 the liquid becomes successively indigo blue, violet, and finally 
 red as more water is added. The light transmitted by such 
 liquids corresponds very closely with that which passes 
 through crystals of the salt containing various proportions of 
 water. Thus, the salts just referred to are known to form 
 definite compounds having a composition expressed as 
 follows : 
 
 Cupric sulphate. 
 
 CuSO 4 . " . . white. 
 
 CuSO 4 5H 2 O . . blue. 
 
 CuBr 5 
 
 Cupric bromide. 
 
 CuBr 9 2H 9 O 
 
 Cobalt chloride. 
 
 CoCl a . 
 CoCl 2 2H 2 O 
 CoCl 2 4H> 2 O 
 CoCl 2 6H 2 
 
 black, 
 blue. 
 
 blue. 
 
 indigo blue, 
 reddish violeu 
 red. 
 
24 Chemical Philosophy. 
 
 In order to account for these and other phenomena 
 connected with solutions, various theories have been pro- 
 posed. These are divisible into two classes, namely, (i) 
 those which assume that a solution is formed as the result 
 of chemical combination between the solid and the solvent, 
 and (2) those which explain the phenomena by the adhesion 
 of the solid to the liquid, and the mechanical intermixture 
 of the two kinds of molecules. 
 
 The view which seems to conform most nearly to the 
 results of observation is the following : A solid immersed in 
 a liquid receives upon its surface the moving molecules of 
 the liquid, and as a result some of the molecules from its 
 surface may become detached and carried off into the 
 liquid. If the encounter of the liquid molecules with those 
 of the solid are incompetent to remove any of them, the 
 substance is insoluble. Once in the liquid the molecules of 
 the dissolved substance may be supposed to move about 
 after the manner of liquid molecules, either singly or in 
 groups, or in company with some of the molecules of the 
 solvent. According to this, the process of dissolution is 
 analogous to the process of evaporation, and just as in the 
 case of evaporation we conceive that molecules which are 
 near to the liquid surface may get entangled there and re- 
 turn to the liquid state again, so we must believe that, so 
 long as a portion of the solid remains undissolved in the 
 liquid, there is an interchange taking place between the 
 liquid and solid molecules near its surface. 
 
 If temperature is raised the process of dissolution is 
 usually promoted, for the molecules of both liquid and solid 
 are more strongly agitated, and the cohesion between the 
 particles of the solid is diminished. But for every given 
 temperature there is a limit to the amount of solid dissolved 
 (in all ordinary cases), when the number of molecules of 
 solid passing off into the liquid no longer exceeds the num- 
 ber of molecules which return from the liquid to the solid 
 surface and are retained there. Change of temperature 
 
Crystallisation. 25 
 
 must be believed to induce a change in the constitution of 
 the liquid itself. In water near the freezing-point the mole- 
 cules probably move about very sluggishly, and very few, if 
 any of them, possess any degree of independence. But if 
 the liquid is heated increasing numbers acquire the sort of 
 movement which is proper to gas molecules, and increasing 
 numbers are ready to pass off from the surface of the mass 
 if they have an opportunity. If the water holds a substance 
 in solution, the molecules of this substance, like those of the 
 solvent, tend to aggregate together at lower temperatures, 
 and to separate from one another and from the molecules 
 of solvent at higher temperatures. 
 
 The question arises, what is it which causes salt mole- 
 cules to unite with water molecules in a saline solution ? 
 The partisans of one view say it is chemical attraction or 
 affinity ; the partisans of the other prefer to call it adhesion. 
 Whatever name be given to it, however, it differs in no 
 obvious quality from the * attraction ' which subsists between 
 the salt and the water in solids which contain * water of 
 crystallisation. 5 (See also Chapter XXII.) 
 
 Naturally connected with the production of solutions is 
 the process of crystallisation, as it is in the gradual passage 
 from the liquid to the solid state that we find the most 
 generally useful method for the formation of crystals. 
 
 The crystallisation of a crystallisable solid from its solu- 
 tions occurs when the percentage of solid present in the 
 liquid exceeds a certain limit dependent on the nature both 
 of the solid and the liquid in which it is dissolved. This 
 condition may be brought about either by allowing part of 
 the solvent to evaporate, or by altering, in general by lower- 
 ing, the temperature. Crystals are then formed, and these 
 crystals, in the majority of cases, contain not only the ele- 
 ments of the dissolved substance, but a portion of the 
 solvent united to it in definite molecular proportions. This 
 is particularly noticeable in the case of aqueous solutions, 
 and the water thus combined is spoken of as water of cry s- 
 
26 Chemical Philosophy. 
 
 tallisation. Alcohol and benzene, and probably other 
 liquids, unite with salts and other crystallisable bodies in 
 the same manner. It is noteworthy that the proportion of 
 water of crystallisation is principally dependent upon the 
 temperature at which the process of crystallisation takes 
 place. Thus, sulphate of sodium crystallises from water at 
 temperatures above 34 in the anhydrous state. But at the 
 ordinary temperature of the air, the solution deposits crys- 
 tals which contain ten molecules of water with one molecule 
 of the salt. 
 
 Certain solutions refuse to crystallise when cooled down 
 below the temperature at which, under ordinary circum- 
 stances, they yield crystals. Such liquids are said to be 
 supersaturated. In order to prepare a supersaturated solu- 
 tion, it is only necessary to warm some water gently in a 
 test-tube or flask and add to it crystals of sodium sulphate, 
 or of alum, or sodium carbonate, so long as they are dis- 
 solved. If the solution is then filtered clear, covered to 
 protect it from dust, and allowed to cool, it will generally 
 remain liquid. If now a small particle of the same salt be 
 dropped into the liquid, crystallisation at once commences 
 and proceeds so rapidly that in a few minutes the whole 
 sets into a solid mass. In most cases crystallisation is not 
 started by shaking the liquid, nor by the introduction of a 
 crystal of a different salt, unless it be strictly isomorphous 
 with the dissolved substance. 
 
 Solution of Gases. All gases dissolve to a greater or less 
 extent in water, but, unlike solids, their solubility diminishes 
 as the temperature rises, so that in most cases * the dissolved 
 gas may be completely expelled from a liquid by boiling, 
 whilst the amount taken up .may be greatly increased by 
 cooling the liquid. 
 
 Increase of pressure also augments the solubility of gases 
 in a direct ratio. It is therefore necessary in making any 
 
 1 Exceptions occur in the cases of hydrochloric acid and some other 
 gases. 
 
Absorption of Gases. 27 
 
 statement as to the solubility of a gas to observe the condi- 
 tions of temperature and pressure under which that solubility 
 was estimated. The following examples will serve to show 
 how greatly gases differ in the extent to which they dissolve 
 in water. 
 
 At o C. and under a pressure of 760 mm. barom. 
 i volume of water will dissolve 
 
 Hydrogen ' OI 93 
 
 Nitrogen '02035 
 
 Oxygen -04114 
 
 Nitrous oxide .... 1*3052 
 
 Carbon dioxide . . . 1*7967 
 
 Hydrogen sulphide . . . 4*3706 
 
 Sulphur dioxide . . . 79*789 
 
 Ammonia 1148-8 
 
 The numbers given above represent volumes of the 
 several gases measured at o and 760 mm., and constitute 
 the coefficients of absorption of these gases at that particular 
 temperature and pressure. 
 
 The general statement that the weight of a gas dissolved 
 by a liquid is directly proportionate to the pressure is often 
 known as the law of Henry and Dalton. It admits of an- 
 other expression ; for since, according to the law of Boyle, 
 the volume of a gas diminishes as the pressure upon it 
 increases, it is obvious that the volume of gas thus held in 
 solution must always be the same, whatever the pressure. 
 
 These rules no longer hold good when the gas and the 
 liquid exert a chemical action upon each other, and excep- 
 tions must also be recognised in the case of the more 
 soluble gases, such as hydrochloric acid and ammonia. 
 
 The determination of the absorption of gases by liquids 
 may be applied in certain cases to the elucidation of some 
 important theoretical and practical questions. Atmospheric 
 air furnishes an example which will be worth the considera- 
 tion of the student. 
 
28 Chemical Philosophy. 
 
 When a mixture of gases is exposed to the action of a 
 solvent, the quantity of each of the constituents dissolved 
 by the liquid will depend first upon its coefficient of 
 solubility, and secondly upon the proportion in which it 
 exists in the mixture. This proportion determines the 
 pressure which each gas present exerts upon the surface 
 of the liquid, and consequently regulates the amount of it 
 which is dissolved. The total pressure produced by the 
 mixture is therefore the sum of those partial pressures due 
 separately to the individual constituents. To make this 
 more clear : 
 
 Suppose a very large vessel containing a very little water, 
 and filled with oxygen, under a pressure of one atmosphere. 
 It is plain that if four-fifths of the gas were removed the 
 pressure would be reduced to one-fifth of an atmosphere, 
 and the quantity of oxygen dissolved would be only one- 
 fifth the quantity taken up under the previous conditions, 
 provided, of course, that the temperature remain constant. 
 An exactly similar vessel can be conceived filled with 
 nitrogen under one atmosphere, and containing a little 
 water. If one-fifth of the nitrogen were removed, the pres- 
 sure of the remainder would be only four-fifths of an atmo- 
 sphere, and the quantity dissolved would be reduced to 
 four-fifths. Lastly, a similar vessel, filled with atmospheric 
 air, contains a gas in which the conditions of the two pre- 
 viously supposed experiments are combined. 
 
 Air is composed very nearly of four volumes of nitrogen 
 to one volume of oxygen, and by reason of the greater 
 solubility of oxygen the proportion of the two gases one to 
 the other is found to be disturbed when air is shaken up 
 with water, the dissolved gas being richer in oxygen, the 
 residual air richer in nitrogen, than the original. 
 
 No stronger evidence could be adduced in favour of the 
 view generally held, that in atmospheric air the two main 
 components are not united chemically, but are in a state of 
 intimate mechanical mixture. 
 
Liquid Diffusion. 29 
 
 Note. The following is an example of the kind of problem 
 that might occur in connection with this subject : 
 
 Calculate the percentage composition of the gas which would 
 be dissolved by water exposed in a room full of air containing 
 79 N, 2O'6 O, and *4 of CO 2 in 100 volumes (temp. o and bar. 
 760 mm.). 
 
 Coeff. of sol. for oxygen ... -04 
 nitrogen . . . -02 
 ,, carbon dioxide . . 179 
 
 The pressures are proportional to the volumes of the gases 
 
 present. Therefore the relative quantities dissolved would be : 
 
 Nitrogen 79 x -02 =1-58 
 
 Oxygen 20*6 x -04 = -824 
 
 Carbon dioxide -4 x 179 = 716 
 
 The total quantity . = 3-120 
 
 The percentage composition of the dissolved gas would 
 therefore be : 
 
 Nitrogen 50-6 
 
 Oxygen ...... 26*4 
 
 Carbonic dioxide .... 22*9 
 
 CHAPTER IV. 
 
 LIQUID DIFFUSION AND DIALYSIS. 
 
 AN aqueous solution of sugar or of salt is heavier than 
 water, and may be readily poured through a funnel with a 
 long stem into a glass of water in such a way as to form a 
 separate stratum at the bottom. If the solution is coloured 
 it will soon be noticed that the colour gradually extends 
 upwards through the liquid, until, after a few hours or a few 
 days, according to circumstances, the whole liquid is tinged. 
 But it is not necessary that the liquid should be coloured. 
 The taste, specific gravity, refractive power, or the applica- 
 tion of chemical tests will soon give indications that the 
 solution from below is mixing with the liquid above. This 
 
3O Chemical Philosophy. 
 
 process of spontaneous intermixture is called diffusion. It 
 results from the proper motion of the molecules of the liquid, 
 and cannot be referred to the disturbing influences of changes 
 of temperature. The rapidity with which diffusion of this 
 kind takes place, and the limit of its action, depend very 
 much upon the nature of the liquids employed. 
 
 The power of interdiffusion is by no means universal 
 among liquids, some liquids being, like mercury, oil, and 
 water, quite incapable of mixing together under any circum- 
 stances ; whilst others, such as water and solution of hydro- 
 chloric acid, mingle spontaneously in consequence of very 
 rapid diffusion. We are indebted for nearly all the informa- 
 tion we possess on this subject to the late Professor Graham. 
 His experiments were conducted very nearly in the manner 
 already described at the beginning of the chapter. The 
 glass vessel in which diffusion was allowed to go on was 
 graduated into equal divisions, from the bottom upwards ; 
 and after the introduction of the two liquids the whole was 
 left in a room, the temperature of which was kept as uniform 
 as possible. 
 
 After a time, the liquid occupying successive divisions 
 of the vessel was removed by a small syphon or pipette 
 and analysed, in order to ascertain the extent to which dif- 
 fusion had taken place. In this way a number of conclu- 
 sions were arrived at, amongst which the following are the 
 most important : 
 
 i. Bodies are divisible, as regards their diffusive power, 
 into two classes. Those which diffuse most readily through 
 a given liquid menstruum are, for the most part, crystal- 
 lisable substances, and are termed by Graham crystalloids ; 
 whilst the least diffusible bodies are uncrystallisable, with, in 
 most cases, high molecular weight, and are denominated 
 colloids, 1 from their resemblance to glue, which may be taken 
 as the type of this class. The following list supplies the 
 times of equal diffusion by the substances there named ; and 
 
 glue. 
 
L iqu id Diffusion. 3 1 
 
 it will be seen that albumen and caramel, both of which are 
 uncrystallisable substances of somewhat indefinite compo- 
 sition, are far behind the rest : 
 
 Hydrochloric acid i 
 
 Chloride of sodium .... 2*33 
 
 Sugar . . 7 
 
 Sulphate of magnesiun* . . 7 
 
 Albumen ...... 1.9 
 
 Caramel ...... 98 
 
 Thus hydrochloric acid diffuses more than twice as rapidly 
 as chloride of sodium, seven times as rapidly as sugar or 
 magnesium sulphate, forty-nine times as rapidly as albumen, 
 and nearly one hundred times as rapidly as caramel. 
 Hydrochloric acid is one of the most diffusive substances 
 known. 
 
 2. Equal rates of diffusion are exhibited in many cases 
 by the members of isomorphous groups. Thus hydrochloric, 
 hydrobromic, and hydriodic acids have nearly the same 
 diffusion rate ; so also have the chlorides, bromides, and 
 iodides of the alkali metals ; the nitrates of barium, 
 strontium, and calcium, and the sulphates of magnesium 
 and zinc. 
 
 3. The rate of diffusion increases with the temperature, 
 and when the solution is not too concentrated is propor- 
 tional to the strength of the solution. 
 
 By taking advantage of this difference in diffusibility, 
 mixed salts may be separated from one another to a certain 
 extent, and crystalloids may be isolated pretty perfectly 
 from admixture with colloids. In the practical application 
 of this process it has been found convenient to separate the 
 liquids undergoing diffusion by some membrane or partition 
 composed of colloid material, and this mode of diffusion, 
 through a septum, is called dialysis. The process is a very 
 simple one. The liquid holding in solution a mixture of 
 crystalloids and colloids is placed in a sort of tray or sieve, 
 
32 Chemical Philosophy. 
 
 formed of a sheet of parchment paper stretched over a hoop. 
 This vessel, which is called the dialyser, is made to float in 
 a dish of pure water, which after a time can be renewed if 
 necessary. Under these circumstances, the crystalloids 
 pass out by diffusion through the membrane (which must 
 be perfectly free from holes), and by evaporating the liquid 
 down may be obtained in a condition of tolerable purity. 
 
 The application of this method led to the discovery of 
 the soluble colloidal forms of ferric hydrate, silicic acid, 
 alumina, and other bodies which had been previously known 
 only in the pectous or gelatinoid condition, and the study 
 of which could not fail to throw considerable light on the 
 obscure natural processes by which these bodies are 
 deposited as minerals in the crystalline form, and in a great 
 state of purity. If, for example, we take a solution of sili- 
 cate of sodium, and add to it a slight excess of hydrochloric 
 acid, we obtain a perfectly clear liquid, which contains the 
 very substance referred to above as colloid silicic acid ; but 
 in this liquid it is mixed with the acid used and the common 
 salt formed by the decomposition. The process of dialysis 
 furnishes the means of separating these latter substances 
 without causing the precipitation or other alteration of the 
 silicic acid, which is left on the dialyser in the form of a 
 colourless limpid solution. This solution is, however, very 
 unstable, especially when concentrated, and the addition of 
 even very minute quantities of various salts causes the whole 
 of the silica to separate out in the form of a translucent 
 jelly, which cannot be re-dissolved, except by the addition 
 of a fresh quantity of alkali. 
 
 Liquid stannic, titanic, tungstic, and molybdic acids 
 have been prepared by a similar process. 
 
 The ultimate pectisation of liquid silicic acid and other 
 colloids is preceded by a gradual thickening of the liquid, 
 and just before gelatinising silicic acid flows like an oil. 
 These effects are doubtless the result of the tendency of 
 the particles of colloids to cohere, aggregate, and contract. 
 
Evaporation and Ebullition. 33 
 
 This tendency manifests itself occasionally in the exercise 
 of very considerable force. Thus the contraction of gela- 
 tine drying in a glass dish over sulphuric acid, together with 
 the adhesion of the gelatine to the glass, is said to be 
 sufficiently powerful to tear up the surface of the glass. 
 Glass is itself a colloid, and the permanent adhesion between 
 the surfaces of polished plates of glass left in contact with 
 each other, which is said to occur sometimes, is referred by 
 Graham to this class of phenomena. 
 
 CHAPTER V. 
 % 
 
 EVAPORATION AND EBULLITION. 
 
 A LIQUID boils when, by raising its temperature, the 
 elasticity of the vapour formed at any point in the liquid 
 is capable of overcoming the pressure at that point. This 
 pressure is made up of the pressure of the atmosphere 
 and that of the superincumbent stratum of liquid. In a 
 vessel of inappreciable depth, and when the mercurial 
 column in the barometer is 760 mm. high, alcohol boils at 
 7$*4 and water at 100. 
 
 The tension of alcohol vapour is, therefore, at any tem- 
 perature below its boiling point, greater than that of water, 
 and alcohol is said to be more volatile than water. Nearly 
 all liquids are volatile, but the temperatures at which they 
 evaporate freely are very diverse. There is, for example, 
 a wide range between the volatility of ether and of molten 
 silver, or between that of liquid carbon dioxide and of 
 metallic mercury. 
 
 The observation of the boiling point of liquids is an 
 operation of daily occurrence in the laboratory, and 
 although very few general laws connecting the boiling 
 point with chemical characters have been traced out, a few 
 
 D 
 
34 Chemical Philosophy. 
 
 general observations have been made, to which the atten- 
 tion of the student must be directed. The conversion of a 
 liquid into a gas or vapour is attended by the absorption of 
 heat, and this heat is consumed in giving to the molecules 
 of the liquid a new and more rapid motion. From this 
 consideration it would appear that bodies formed of light 
 molecules would be more easily vaporised than others con- 
 stituted of complex and consequently heavy molecules. 
 Such is indeed the case, and, in the broadest sense, such a 
 statement would be nearly true. But when we examine 
 individual cases, we meet with so many exceptions and 
 anomalies, that it is obvious such a law must be applied 
 with extreme caution. It is, however, permissible to say 
 that bodies which are strictly comparable in regard to 
 chemical and other physical qualities do in nearly all cases 
 exhibit the relation referred to. Among the elements the 
 most noteworthy instances are the following : 
 
 Molecular Weight. Boiling Point. 
 
 Chlorine . . 71 ~5 
 
 Bromine. . .160 63 
 
 Iodine , . . 254 175 
 
 Oxygen . . .32 gaseous 
 
 Sulphur . . .64? 440 
 
 Selenion . . . 159? below a red heat 
 
 Tellurium . . 252? white heat 
 
 Nitrogen . . 28 gaseous 
 
 Phosphorus . .124 290 
 
 Arsenic , . .300 volatilises at red 
 
 heat without fusion 
 
 Antimony . . 240 ? white heat 
 
 Bismuth . . 416? white heat 
 
 Groups of similarly constituted compounds show the 
 same relations. 
 
Boiling Points. 35 
 
 Molecular Weight. Boiling Point. 
 
 Sulphur Dioxide .64 - 10 
 
 Sulphur Trioxide . 80 46 
 
 Chloromethane . 50-5 gaseous at ordinary 
 
 temperatures 
 
 Dichloromethane . 85 3'5 
 
 Trichloromethane .119*5 61 
 
 But the most important cases are observable among the 
 carbon compounds, which form homologous series, to which 
 reference will be made hereafter. One example will suffice 
 in this place. The following compounds all contain carbon 
 and hydrogen in proportions which may be represented by 
 the general formula C n H 2n+2 : 
 
 Molecular Weight. Boiling Point. 
 
 Methane (marsh gas) 16 gaseous 
 
 Ethane ... 30 gaseous 
 
 Propane ... 44 gaseous 
 
 Tetrane ... 58 i 
 
 Pentane ... 72 38 
 
 Hexane ... 86 70 
 
 Heptane . . .100 99 
 
 Octane ... 114 124 
 
 &c. &c. 
 
 In this series, for every increase of 14 on the molecular 
 weight, the boiling point rises by about 30 to 35. 
 
 In the determination of boiling points it is usual to 
 observe, by an accurate thermometer, the temperature of the 
 vapour evolved by the boiling liquid, and not that of the liquid 
 itself. This precaution is necessary when glass vessels are, 
 as usual, employed, in consequence of a peculiar adhesive 
 attraction which glass exercises, and which causes the boil- 
 ing point to be slightly raised above the true temperature 
 of ebullition. This adhesion, assisted, doubtless, by cohesion 
 
 D 2 
 
36 Chemical Philosophy. 
 
 in the liquid itself, sometimes gives rise to the phenomenon 
 of irregular ebullition or * bumping.' The entire thread of 
 mercury in the thermometer should be immersed in the 
 vapour, so as to be heated by it, and the height of the 
 barometer should be noted at the time of the experiment. 
 
 This last is a precaution which is very generally 
 neglected, and its neglect is probably the cause of some 
 of the discrepancies noticed between calculated and 
 observed boiling points and between the results of different 
 experiments. 
 
 The determination of the boiling point is often useful in 
 deciding as to whether a given liquid is a mixture or a 
 homogeneous body. When heat is applied to a mixture of 
 volatile liquids the mixture begins to boil at a temperature 
 very near to the boiling point of its most volatile con- 
 stituent, and if the temperature is not allowed to rise above 
 this point, ebullition in most cases soon comes to an end. 
 But if the application of heat is continued whilst the thermo- 
 meter is kept in the vapour, the temperature may be observed 
 to rise continuously till the whole of the liquid has boiled 
 away. If this operation is conducted in a flask connected 
 with a condensing apparatus and a receiver, and if the 
 receiver is changed at intervals, so that the several portions 
 which pass over between certain limits of temperature are 
 received in separate vessels, a more or less complete separa- 
 tion of the constituents of the liquid may be effected. Such 
 a process is called fractional distillation. 
 
 The compression of a vapour tends to produce the same 
 change of state as lowering its temperature. In either case 
 the approximation of the molecules is attended sooner or 
 later by the liquefaction of a part of the vapour. With these 
 facts in view, and considering the generally close resemblance 
 between vapours and those bodies which are commonly 
 called true gases, Faraday came to the conclusion that the 
 latter are not essentially different from the former, but are 
 in truth the vapours of volatile liquids far removed at 
 
Liquefaction of Gases. 37 
 
 ordinary temperatures from their boiling points. This con- 
 clusion he verified experimentally by enclosing in strong 
 ^-shaped glass tubes materials capable of evolving the gases 
 he wished to examine. On the application of a gentle heat 
 to these materials gas was generated, and by its accumula- 
 tion in the confined space, sufficient pressure was exerted 
 to cause its partial liquefaction. In this way ammonia, 
 chlorine, and other gases were reduced to the condition of 
 limpid liquids, and by the combined use of pressure and 
 low temperature, produced by powerful freezing mixtures, a 
 great many other bodies, which till then had been known 
 only in the gaseous form, were also liquefied. Larger 
 apparatus, constructed on the same principle as Faraday's 
 glass tubes, were subsequently employed by different experi- 
 menters, and at the present time several gases, such as 
 carbonic anhydride and nitrous oxide, are liquefied on a 
 large scale by compressing them by powerful force-pumps 
 into iron cylindrical bottles fitted with stopcocks. By such 
 methods all known gases were, with six exceptions, reduced 
 to the liquid state. 
 
 It was not until the end of the year 1877 that these ex- 
 ceptions disappeared, and it was found possible by application 
 of enormous pressure assisted by very low temperature to 
 reduce to the liquid state such gases as oxygen, hydrogen, 
 nitrogen, and atmospheric air. This interesting result was 
 arrived at almost simultaneously, though independently, by 
 M. Pictet of Geneva and M. Cailletet of Chatillon-sur-Seine. 
 A temperature of 130 combined with a pressure of about 
 475 atmospheres is required for the liquefaction of oxygen, 
 whilst at about the same temperature a pressure of some 
 650 atmospheres is necessary in the case of hydrogen. 
 
 The evaporation of liquefied gases is attended by the 
 absorption of much heat, and in some cases the reduction 
 of temperature is such as to cause the solidification of part 
 of the liquid. Carbonic anhydride can be obtained in the 
 form of a white snow-like solid by allowing a fine stream of 
 
38 Chemical Philosophy 
 
 the liquefied gas to escape into the air. Part of it evapo- 
 rates very rapidly, and so much heat is thus rendered latent 
 that the remainder freezes. Hydrogen is also said to yield 
 a solid, though in this case, owing to the difficulties attending 
 the experiment, there is some doubt about the observation. 
 The following table shows the amount of pressure in 
 atmospheres necessary at the temperature of o to liquefy 
 some of the more important of the gases : 
 
 PRESSURE IN ATMOSPHERES- 
 
 Sulphur dioxide . . . . 1*53 
 
 Cyanogen 2-37 
 
 Hydriodic acid . . . . 3-97 
 Ammonia . . . . . 4*4 
 Chlorine .... about 4 
 Hydrogen sulphide . . .10 
 
 Nitrous oxide . . . . 32 
 
 Carbon dioxide .... 38*5 
 Hydrochloric acid , , . about 42 
 
 Some of these may be liquefied by cold alone under the 
 ordinary atmospheric pressure. Thus 
 
 Sulphur dioxide condenses at . , 10 
 
 Cyanogen . . . , 22 
 
 Ammonia . . . 36 
 
 Chlorine . . . . 50 
 
 Carbon dioxide . . . . 87 
 
 On the other hand, Andrews found (see also p. 10) that 
 if the temperature be raised to a certain point, a gas which is 
 otherwise liquefiable can no longer be liquefied by pressure. 
 This point, which varies with the nature of the gas operated 
 upon, is called the critical point. In the case of carbon 
 dioxide it is 3 1 C. The critical points for hydrogen, oxygen, 
 and the rest of the gases which were formerly supposed to 
 
Liquefaction of Gases. 39 
 
 be permanent are at temperatures so low as to be attainable 
 only by special arrangements and with great difficulty. 
 Hence the failure of the earlier attempts to reduce these 
 gases to the liquid state, notwithstanding the application of 
 pressures represented by several thousand atmospheres. 
 
 It has been already stated earlier in the chapter that if 
 we compare together liquids which consist of the same 
 elements, and which present the same general properties, 
 those which have the simplest constitution are the most 
 readily converted into vapour. The converse is also, to a 
 certain extent, true. Comparing together similarly consti- 
 tuted gases and vapours, we find that those which are com- 
 posed of simple molecules are often more difficult to liquefy 
 by cold or compression than others of more complex consti- 
 tution. These facts are sometimes serviceable in helping to 
 decide questions as to the relative complexity of two nearly 
 allied compounds. For example, there are two oxides of 
 carbon, both gaseous at ordinary temperatures, but one of 
 them more easily capable of liquefaction under pressure 
 than the other. The more easily liquefiable oxide is in all 
 probability made up of heavier molecules than the other, 
 and this view is supported by a comparison of the densities 
 of the two gases. Carbonic anhydride gas is, bulk for bulk, 
 i '5 7 time heavier than carbonic oxide. 
 
 Nitrous and nitric oxide, ethylene and marsh gas, furnish 
 examples of the same kind. 
 
 Such a rule, however, is liable to many exceptions, and 
 can only be used in conjunction with other considerations. 
 Thus ether, C 4 H 10 O, is more volatile than alcohol, C 2 H 6 O. 
 Moreover, it often happens that the same gas or vapour 
 changes in constitution at different temperatures. For ex- 
 ample, nitric peroxide above 180 is composed of molecules 
 whose formula is NO 2 . At lower temperatures these com- 
 bine in pairs to form more complex molecules, N 2 O 4 , so that 
 when the gas is about to liquefy it consists of molecules of 
 greater mass than when it is hot. Hydrofluoric acid affords 
 
4O Chemical Philosophy. 
 
 another example of the same kind. Its molecule consists 
 of HF at moderately high temperatures, whilst at a little 
 above 20 it becomes H 2 F 2 . It is probable that there are 
 many cases of this kind. (See Chapter XVIIL, ' Disso- 
 ciation.') 
 
 CHAPTER VI. 
 
 DIFFUSION AND DIALYSIS OF GASES 
 
 A VERY remarkable property of gases and vapours is their 
 power of mixing with one another, even in opposition to 
 gravity. 
 
 If a bottle of any odorous gas is opened in any part of a 
 room of constant temperature and free from draughts, the 
 smell of the gas soon becomes perceptible in every part 
 of the room, and, after the lapse of a short time, equally 
 in every part. Notwithstanding that, as in the case of 
 sulphuretted hydrogen, a heavy gas may be selected for the 
 experiment, it would be easy to prove by analysis of the air 
 that every part of it is equally impregnated with the foreign 
 matter. Other experimental illustrations of the same law 
 may easily be devised. A bottle of hydrogen hetd mouth 
 downwards in the air for a short time soon becomes ex- 
 plosible. A jar of air inverted over another filled with 
 carbonic acid gas soon acquires the power of giving, like 
 carbonic acid, a precipitate with lime water. And this 
 process of intermixture proceeds almost equally well if the 
 gases are separated from each other by a partition formed 
 of some porous material. 
 
 A thin plate of unglazed earthenware, a slice of artifi- 
 cially compressed graphite, or a cake of dry plaster of Paris, 
 may be employed for the purpose. A very effective form of 
 apparatus consists of a clay battery cell closed by a cork, 
 through which a yard or so of glass tubing open at both 
 
Law of Gaseous Diffusion. 41 
 
 ends is made to pass. By means of this simple apparatus 
 it may be shown that different gases penetrate the porous 
 clay with different degrees of rapidity, and that light gases 
 effect a passage more quickly than heavier ones. The con- 
 sequence of this difference of diffusion-rate is that a differ- 
 ence of pressure is established inside the cell, and if the 
 open end of the tube is dipped into water a certain quantity 
 of gas is expelled in bubbles from below, or the liquid is 
 forced by the atmospheric pressure up the tube. If, for 
 example, the clay vessel previously full of air is surrounded 
 by hydrogen gas, intermixture of the air within and the 
 hydrogen without takes place through the clay ; but since 
 the hydrogen diffuses more rapidly than the air, the quantity 
 of gas within is rapidly increased, and some of it visibly finds 
 its escape from the open end of the tube through the water. 
 Early observations of and experiments upon gaseous 
 diffusion were made by Priestley in the last century, and 
 by Dobereiner in 1825 ; but Graham gave the explanation 
 of the phenomena, and by precise and long-continued ex- 
 periments established the law : 
 
 The velocities of diffusion of different gases are inversely 
 proportional to the square roots of their densities. 
 
 Graham's experiments were for the most part conducted 
 with a very simple apparatus, consisting of a straight wide 
 glass tube closed at its upper extremity by a disc of porous 
 stucco or graphite. This tube was filled with hydrogen or 
 other gas over the mercurial trough, the graphite plate being 
 covered during this operation with a sheet of guttapercha. 
 The mercury within the tube was kept at the same level as 
 the mercury in the trough, in order that there might be no 
 alteration of pressure whilst the diffusion was proceeding, 
 and at the same time the temperature and barometric 
 pressure were recorded. After the lapse of a certain 
 interval, measured by a chronometer, the volume and com- 
 position of the residual gas could be determined. 
 
 The following table embodies some of the results 
 
Chemical Philosophy. 
 
 obtained in this way, the pressure and temperature being 
 supposed to be the same in all cases : 
 
 DIFFUSION OF GASES. 
 
 Name of Gas 
 
 Density 
 
 Square Root 
 of Density 
 
 i 
 
 ^/Density 
 
 Velocity of 
 Diffusion 
 
 Air 
 
 I 
 
 I 
 
 I 
 
 I 
 
 Hydrogen . 
 
 0693 
 
 2632 
 
 37794 
 
 3-83 
 
 Marsh gas 
 
 '554 
 
 774 
 
 i'3375 
 
 1-344 
 
 Carbonic oxide 
 
 9678 
 
 9837 
 
 1-0165 
 
 I-OI49 
 
 Nitrogen . 
 
 9713 
 
 9856 
 
 1-0147 
 
 I-OI43 
 
 Oxygen . 
 
 1-1056 
 
 i"05i5 
 
 9510 
 
 9487 
 
 Carbonic dioxide 
 
 1-529 
 
 1-2365 
 
 8087 
 
 812 
 
 It will be noticed that the observed rate of diffusion 
 agrees very nearly with the rate calculated from the density 
 of the gas ; but in no case is there absolute concordance. 
 This is, probably, in part due to the errors inevitable in 
 any experimental investigation, especially where gases are 
 concerned, but is also in some degree attributable to the 
 fact that the diaphragm employed possesses an appreciable 
 thickness, so that in passing through the pores the gas 
 encounters considerable resistance. 
 
 By taking advantage of the unequal diffusibility of gases 
 of different density, a partial separation of mixed gases may 
 in some cases be effected. 
 
 The gases constituting atmospheric air, for example, may 
 be, to some extent, separated from each other by causing a 
 slow current of air to flow through a clay tube passing 
 through a glass tube which has been exhausted as com- 
 pletely as possible by the air-pump. The nitrogen, being 
 lighter and consequently more diffusible than the oxygen, 
 passes more abundantly into the vacuous space, leaving the 
 residual air richer in oxygen than it was originally. 
 
 The physical explanation of the phenomena of diffu- 
 sion depends directly upon the mechanical theory of gases, 
 
Passage of Gases through Porous Septa. 43 
 
 which has been (Chap. I.) already discussed. The mole- 
 cules of a gas are supposed to move constantly in straight 
 lines till they come nearly into contact with other moving 
 molecules, or with the walls of the containing vessel. If 
 these walls are perforated at intervals with apertures large 
 enough to permit the passage of a molecule, we may con- 
 ceive that although many molecules continue to rebound 
 as though the surface were impervious, yet a great many 
 others may find their way into and through these short pas- 
 sages, and so into the atmosphere beyond. Molecules from 
 the external atmosphere may be assumed to pass inwards in 
 precisely the same manner, and if the densities of the 
 gases on the two sides of the partition are the same, the 
 number of molecules passing inwards in a given time is 
 exactly equal to the number passing outwards, and no 
 change of volume or of pressure can result. But if the 
 gases are of different densities, molecules of the lighter gas 
 pass through more rapidly than those of the heavier, and a 
 change is produced in the tension or elastic force of the gas 
 enclosed in the porous vessel. 
 
 This pressure, exerted by a gas in opposition to that 
 which it has to bear when enclosed in a vessel, is represented 
 according to the kinetic theory as the result of the con- 
 tinuous showering down of its molecules upon the surfaces 
 with which it is in contact. The molecules of different gases 
 being of different masses, they must move with different 
 degrees of velocity, the light molecules more rapidly than 
 heavier ones, in order to produce the same amount of pres- 
 sure. Hence, when the gas in contact with the porous sur- 
 face is a light gas, its molecules must be supposed to fall 
 upon a given area more frequently than when the gas em- 
 ployed is heavier, and consequently the opportunities for 
 the escape of molecules through the pores are more frequent. 
 Hence it is that light gases diffuse more rapidly than heavy 
 gases. 
 
 Gases have riot only the power of passing by diffusion 
 
44 Chemical Philosophy. 
 
 through porous substances, but under certain circumstances 
 penetrate membranes, and even sheets of metal which are 
 absolutely destitute of pores. This phenomenon differs 
 entirely from diffusion, for it is not found that the lightest 
 gases traverse such substances most rapidly ; indeed, the 
 contrary is more generally the case. Moreover, the metals 
 which are so remarkable for their power of transmitting 
 some gases are absolutely impermeable by others. 
 
 A few simple experiments will give the student an 
 idea of the general character of the phenomena we are dis- 
 cussing. If a thin india-rubber balloon (such as are sold at 
 the toy-shops), inflated with air, is immersed for a few 
 minutes in a vessel full of carbonic dioxide gas, the balloon 
 becomes largely distended, and if a band of tape is fastened 
 round it before the experiment, it generally bursts after im- 
 mersion in the gas for a short time. A similar balloon filled 
 with hydrogen or carbonic dioxide gas quickly collapses 
 when exposed to the air. Such a film of rubber appears to 
 have no porosity, but rather to resemble a film of liquid in 
 its relations to gases. The penetration of the rubber and 
 similar colloids by a gas appears to be due to the absorption 
 of the gas by one surface of the colloid and its transmission 
 to the other surface by the agency of liquid and not gaseous 
 diffusion. The liquefied gas then volatilises into the vacuum 
 or atmosphere on the other side. 
 
 The passage of gases through metallic plates at a red 
 heat is referred by Graham to a somewhat similar cause. 
 Thus, at a red heat, both platinum and palladium, and even 
 iron, are permeable by hydrogen gas ; and this is evidently 
 connected with the fact that the same metals are capable of 
 absorbing and retaining considerable quantities of hydrogen 
 when that element is presented to them under suitable con- 
 ditions. Thus a sheet of palladium connected with the 
 negative pole of a battery, and immersed in acidulated water, 
 becomes charged with upwards of 200 times its volume of 
 hydrogen gas. The same metal in a spongy state absorbs 
 
Condensation of Gases by Solids. 45 
 
 686 times its volume of hydrogen when heated in the gas to 
 200. This ' occlusion ' of hydrogen is not attended by 
 any alteration in the appearance of the metal, although its 
 volume is increased, and consequently its density diminished. 
 
 Other gases are occluded in a similar manner by other 
 metals ; but in each case a certain selective power is mani- 
 fested on the part of the metal. Thus, platinum and pal- 
 ladium take up hydrogen freely, but no other gas to an 
 appreciable amount ; iron takes up hydrogen and carbonic 
 oxide, and melted silver absorbs oxygen. 
 
 It is not improbable that these remarkable effects may 
 be special manifestations of the absorptive power which is 
 possessed in greater or less degree by the surfaces of all 
 solids. Even glass exposed to air condenses a quantity of 
 gas which extends, in the form of an invisible and almost 
 imponderable film, over the entire surface. The gas thus 
 collected from air by glass consists of the same ingredients 
 as are found in common air, namely, oxygen, nitrogen, 
 carbon dioxide, and water; but it is not quite certain whether 
 they are present in the same proportions. If the surface is 
 greatly extended, as is the case with porous substances like 
 plaster, earth, and especially charcoal, the amount of gas 
 taken up is often very large, and each substance seems to 
 possess a power of selection from the gases presented to it. 
 Cocoa-nut charcoal, for example, at the common tempera- 
 ture and pressure of the air, absorbs about 170 times its bulk 
 of ammonia, but only about 73 times its bulk of carbon 
 dioxide, and a very much smaller quantity of oxygen, nitro- 
 gen, or hydrogen. 
 
 As a general rule those gases which are easily liquefi able 
 are absorbed by such substances in largest quantity. 
 
46 Chemical Philosophy. 
 
 CHAPTER VII. 
 
 RELATION OF GASES TO TEMPERATURE AND PRESSURE. 
 
 THE bulk of a given portion of matter, solid, liquid, or 
 gaseous, must depend both upon the temperature and the 
 pressure to which it is submitted. The variation of volume 
 produced in a solid or liquid by change of pressure or of 
 temperature is very small when compared with the change 
 produced by the same means in the volume of a gas. The 
 law of the compressibility of gases was discovered by the 
 Hon. Robert Boyle and published in 1662. It was examined 
 independently and verified by Mariotte some years after- 
 wards. 
 
 Law of Boyle, * The volume of a given mass of any gas 
 varies inversely as the pressure.' 
 
 Thus, if V is the volume when the pressure is P, 
 
 The volume of the gas becomes 
 
 \ V when the pressure is . . 2 P 
 I V " P 
 
 3 v 35 O - 1 - 
 
 iv . . 4 P 
 
 JV . . n P 
 
 Also 
 
 2V . . JP 
 
 3V . ., IP 
 
 V J P 
 
 n V - f 
 
 Hence the pressure which is produced by the elastic force 
 or tension of a gas is proportional to its density. 
 
 So that if pressure increases, 
 
 density increases, 
 and volume diminishes. 
 Abo, if volume increases, 
 
 pressure and density diminish. 
 
Laiv of Boyle. 47 
 
 Boyle's law is not absolutely obeyed by any known gas ; 
 but hydrogen and the other gases which are liquefiable with 
 great difficulty conform to the law very nearly, and thus 
 present the nearest approach to the condition of a perfect 
 gas with which we are acquainted. With change of pressure 
 the easily liquefiable gases and vapours increase or decrease 
 in volume to a greater extent than more permanent gases. 
 
 The pressure and density of atmospheric air, and of 
 gases which are in communication with it, are estimated by 
 the aid of the barometer. This instrument, in its simplest 
 form, consists of a straight glass tube, somewhat less than 
 a meter long, and closed at one end. The tube is filled 
 with pure mercury, free from air, and then inverted with 
 the open end beneath the surface of pure mercury. The 
 liquid metal then falls from the closed extremity, leaving a 
 space which is generally referred to as the 'Torricellian 
 vacuum.' It contains nothing but mercurial vapour. It 
 is usual to consider that the atmosphere possesses its 
 average and normal density when, at the sea level and at the 
 temperature of o C., the column of mercury sustained by 
 the atmospheric pressure is 760 millimetres (or 29-92 
 inches) high, measuring from the surface of the mercury 
 in the reservoir to the surface of the mercury within the 
 tube. This amount of pressure is often spoken of as one 
 atmosphere. 
 
 In accordance with the law of Boyle, the volume of a 
 gas under altered barometric pressure can be calculated by 
 the formula 
 
 V_P, 
 
 v; p ' 
 
 in which V is the given volume under pressure P, and V\ 
 is the new volume when the pressure is altered to PI. So 
 that 
 
 Example. 100 volumes of air are measured off when 
 
48 Chemical Philosophy. 
 
 the barometric pressure is 740 mm. ; what will be the 
 volume of the same air when the barometer stands at 
 760 mm. ? 
 
 Here V = 100, P = 740, P t = 760. 
 
 Then V, = I0 * 74 = 97-3 vols. 
 700 
 
 Answer. 
 
 LAW OF GAY-LUSSAC AND CHARLES. 
 
 Air expands by of its volume at o for every increase 
 in temperature of i C. 
 
 Thus 273 volumes of air at o 
 
 become 274 ,, i 
 
 275 2 
 
 276 ... 3 
 273 + * . . t 
 
 Also, 273 . o 
 
 become 272 . . 1 
 
 271 . -2 
 
 270 - 3 
 
 273-t -t 
 
 And generally 273 + 1 at t become 273 + T at T. 
 
 This fraction , or -003665, is called the coefficient of 
 
 273 
 
 expansion, and represents almost exactly the increment or 
 decrement which occurs in a measured volume of air or 
 other permanent gas for every change of temperature of one 
 degree centigrade, provided the pressure remains unchanged. 
 The coefficient of all gases is very nearly coincident with 
 that of air, and for chemical purposes may, without incon- 
 venience, be assumed to be the same. Strictly speaking, 
 however, every gas has a coefficient of its own, which, in 
 the case of the more easily liquefiable gases, is perceptibly 
 
Expansion of Gases by Heat. 49 
 
 greater than the number given above, as may be seen by 
 the following table : 
 
 Coefficients of expansion for i C. 
 
 Air . . . . . '003665 
 
 Hydrogen .... -003667 
 
 Carbonic oxide . . . -003667 
 
 Nitrogen .... -003668 
 
 Nitrous oxide . . . -003676 
 
 Carbonic anhydride . . -003688 
 
 Cyanogen .... -003829 
 
 Sulphurous anhydride . -003845 
 
 It seems not unreasonable to suppose that such differ- 
 ences are due in part to the fact that in vapours and in the 
 more easily liquefiable gases the influence of cohesion is not 
 altogether annulled. It is conceivable that a vapour may 
 consist of molecules which, unlike the independent mole- 
 cules of perfect gases, may be connected together into com- 
 panies moving about much in the same way as individual 
 molecules, but less rapidly. 
 
 Examples. A certain mass of air measures 100 cubic 
 centimetres at o; to find its volume at 10. 
 
 273 vols. of a gas at o become 273 + 1 vols. when the 
 temp, is t C. 
 
 In this case t = 10. 
 
 Then 273 c.c. at o become 283 c.c. at 10, 
 
 i becomes ^-^ c.c. at 10, 
 
 273 
 and ioo,, 283 x IOQ cc ^ IQO 
 
 273 
 Ans. 103*66 c.c. 
 
 Or employing the decimal equivalent to - , let Vi be the 
 
 required volume and V the vol. given. 
 
 E 
 
5O Chemical Philosophy. 
 
 Then Vj = V (i + '00366 t) 
 
 = TOO (l + '00366 X TO) 
 
 = 103-66 Ans. 
 
 300 c.c. of air at 20 ; find the volume at o. 
 273 + 20 at 20 become 273 at o 
 
 i becomes ^ at o. 
 
 293 
 
 And 300 a73 a * 3 3 ato. 
 
 279-5 ^ ns - 
 
 Or, using the decimal coefficient, we say V , a certain 
 volume of air at o, becomes 300 c.c. at 20, or 
 
 V (i + "00366 x 20) = 300. 
 
 Whence V == - - - = 279-5 c - c - <A- ns - 
 i + -00366 x 20 
 
 500 c.c. of air at 10 ; find the volume at 10. 
 
 In this and all similar problems it is to be remembered 
 that the coefficient of expansion is a fraction of the volume 
 which the gas occupies at o, not at any other temperature, 
 
 273 + 10 c.c, of air at 10 
 
 measure 
 273 10 c.c. at 10. 
 
 Therefore 500 c.c. of air at 10 
 
 measure 3 * 5 , or 464-6 c.c. at 10. 
 283 
 
 Or let YIO, V , and V , be the volumes at the tem- 
 peratures 10, o, and 10 respectively ; then 
 
 V-j = V (i - -00366 x 10) = - '- - x (i - -00366 x 10) 
 
 I + -00366 X 10 
 
 = 464-6 c.c. Ans. 
 
 Air or any other permanent gas diminishes by -- of 
 its volume for every degree of temperature travelling down 
 
Absolute Temperatures. 5 I 
 
 the scale. If the same relations of volume to temperature 
 were maintained, it is obvious that at 273 the volume 
 would be /7, and the gas cease to exist. Such a tempera- 
 ture has, however, never been attained ; and if ever such a 
 degree of cold were reached, there can be no doubt that a 
 gas exposed to it would liquefy, or that some change would 
 occur whereby the gas would be released from obedience to 
 the ordinary law. 
 
 Notwithstanding, however, that such a condition of things 
 is practically beyond the reach of experiment, this con- 
 sideration is important as furnishing the basis of an abso- 
 lute scale of temperature. Calling 273 C. the zero point, 
 we represent absolute temperatures by adding 273 to the 
 number of degrees upon the ordinary Centigrade scale. 
 From what has already been stated regarding the expansion 
 of gases, it follows that, pressure being constant, the volume 
 of a mass of gas varies directly as the absolute temperature. 
 This statement is sometimes referred to as the law of 
 Charles, to whom we owe the discovery * of the equal ex- 
 pansibility of the principal gases by heat. 
 
 LAW OF AVOGADRO. 
 
 It has been shown in the foregoing paragraphs that all 
 gases, when under conditions sufficiently remote from those 
 which induce their liquefaction, are affected in the same 
 manner and to the same extent ly changes of pressure 
 and of temperature. Differences of density, of chemical 
 composition, or of chemical properties do not affect the 
 generality of this statement. The volumes of heavy oxygen 
 and light hydrogen, of simple nitrogen and compound 
 marsh gas, increase and decrease according to the same law. 
 It is impossible to avoid the inference from these facts that 
 these gases, so different chemically, must be physically con- 
 
 * Towards the end of the last century. 
 
 E 2 
 
52 Chemical Philosophy. 
 
 stituted alike. If now we admit the hypothesis that gases, 
 like other bodies, are made up of small independent masses 
 called molecules, and that heat causes these molecules to 
 separate from one another, whilst cold or pressure causes 
 them to approach, we are led to the assumption that in 
 equal volumes of different gases l there must exist the same 
 number of molecules. 
 
 This statement, originally enunciated by an Italian 
 physicist, Avogadro, in 1811, may now be regarded as a 
 well-established truth. 
 
 But, like every other part of the molecular theory, this 
 law owes its recognition by physicists and chemists, not to 
 any direct proof that can be adduced from experimental 
 sources in support of such hypothesis, but to the fact that 
 nearly all observed chemical phenomena do not only har- 
 monise with such views but find in them complete and satis- 
 factory explanation. 
 
 Admitting the law of Avogadro, we see at once why 
 gases are equally expanded by heat, why they are equally 
 contracted by cold and pressure, and why they combine 
 together, according to the discovery of Gay-Lussac, in 
 simple proportions by volume. 
 
 In a later chapter will be shown some of the conse- 
 quences which follow upon an application of this law, and 
 the important progress of chemical theory which has resulted 
 from its adoption. 
 
 1 Under the same circumstances of temperature and pressure. 
 
The Spectrum. 53 
 
 CHAPTER VIII. 
 
 SPECTRA EMISSION SPECTRA. 
 
 ALL bodies when heated to a sufficiently high temperature 
 emit light ; those which are densest being, as a rule, the 
 most intensely luminous. 
 
 When this light is examined by a prism the image formed 
 by the refracted and dispersed rays appears in the form of 
 a coloured band, which is called the spectrum. The appa- 
 ratus employed for the purpose of observing the spectra of 
 different kinds of light is called a spectroscope. The details 
 of its construction are described in nearly all works on 
 physics. Suffice it, therefore, to say that the light under 
 examination is allowed to pass first through a fine slit in a 
 metallic plate so arranged that the slit is parallel to the 
 edges of the prism. The rays are then rendered parallel, 
 by means of a pair of lenses placed in a tube which is fixed 
 at the angle of minimum deviation with the first face of the 
 prism. After passing through the prism the light is viewed 
 through a telescope, which gives a magnified image of the 
 spectrum. 
 
 The spectrum of the light emitted by solids, liquids, and 
 very dense gases is usually found to be a continuous one ; 
 that is to say, the simple colours red, orange, yellow, green, 
 blue, indigo, and violet which are its components merge 
 one into the other gradually, and are not separated by 
 dark intervals. The spectra furnished by ordinary gases or 
 vapours, when ignited, consist, on the contrary, of bright 
 lines, which are so many images of the slit of the spectro- 
 scope refracted in different degrees so as to be separated 
 from one another. In the spaces between these bright 
 bands there is in general no light. In some cases, however, 
 a continuous spectrum is more or less distinctly visible. .. 
 
54 Chemical Philosophy. 
 
 These differences are shown as far as possible without the 
 aid of colour in the following diagrams. The first represents 
 the appearance which is presented by the spectrum of an 
 ignited solid, such as lime, or of a flame such as that of a 
 candle, which is supposed to contain either solid matters or 
 very dense vapours. The second is the spectrum of heated 
 hydrogen gas. 
 
 CONTINUOUS SPECTRUM. 
 
 BRIGHT LINE SPECTRUM. 
 
 Salts of the alkalis, of the alkaline earths, of copper and 
 many other metals, have long been known to give light 
 having peculiar and characteristic colours when heated in 
 the blow-pipe flame or in the non-luminous flame of the 
 Bunsen burner. Thus, sodium salts give yellow light; 
 potassium, violet ; barium, green ; calcium, orange-red ; 
 and lithium and strontium, crimson flames. 
 
 When flames coloured by the vapours of these salts are 
 viewed with the' spectroscope, they are found to exhibit a 
 character similar to that of the gas hydrogen. That is to 
 say, these spectra are made up of bright lines which are 
 separated from each other by dark intervals. The appear- 
 ance shown in each case is indicated by the diagrams on 
 next page. 
 
 When the same prism is employed these lines always 
 occupy the same relative positions, and the lines produced 
 by any one substance are not changed in position, or 
 breadth, or intensity in the presence of another substance 
 which, when ignited, gives out a different kind of light. v 
 
Line Spectra. 
 
 55 
 
 VIOLET. 
 
 Thus, if a flame coloured by sodium or one of its salts is 
 viewed through the spectroscope, the bright yellow double 
 band characteristic of sodium is alone visible. 
 
56 Chemical Philosophy. 
 
 Another flame coloured by some compound of potassium 
 gives only the red and deep blue lines peculiar to the 
 ignited vapour of that metal. A third flame, into which is 
 introduced a mixture of potassium and sodium salts, appears 
 yellow to the unaided eye, if the proportion of sodium 
 present is more than infinitesimal. But after passing through 
 the prism of the spectroscope this yellow light is resolved 
 into red and indigo potassium bands and the yellow sodium 
 lines occupying exactly the same position as when observed 
 separately. These facts constitute the basis of the method 
 of spectral analysis. 
 
 It is only necessary to bring before the slit of the 
 spectroscope the incandescent vapour of a metal or other 
 substance which it is desired to examine. The position 
 and number of the bright lines visible through the telescope 
 are at once an indication of the nature of the substance. The 
 volatilisation of solid bodies and the necessary ignition of 
 the vapour is effected most generally by the aid of the 
 Bunsen gas-flame ; but when this is incompetent to produce 
 a temperature sufficiently high, the oxy-hydrogen blow-pipe 
 or the electric arc may be employed. The transmission of 
 sparks from an induction coil between terminals to which 
 metals can be attached is a convenient method for obtaining 
 the spectra of such substances, as well as of gases such as 
 nitrogen, hydrogen, and carbon dioxide, through which the 
 sparks can be passed. 
 
 Another method of obtaining the spectra of gases con- 
 sists in enclosing them in glass tubes into the walls of which 
 metallic wires are sealed, which serve as poles between 
 which the discharge of an induction coil can pass. On 
 exhausting the tubes by the air-pump so as to remove the 
 greater part, but not the whole, of the enclosed gas, the 
 discharge occurs much more readily than in the same gas 
 under atmospheric pressure, and the attenuated gas becomes 
 brilliantly luminous, and the light can be readily examined 
 by the spectroscope. 
 
Spectroscopic Analysis. 57 
 
 The delicacy of the method of spectral observation is 
 very great, far surpassing that of the most exact and sensi- 
 tive of chemical tests, and by its aid the presence of ex- 
 ceedingly minute quantities of various elements can be 
 detected with certainty and ease. Sodium and lithium, for 
 example, are elements which, even in excessively small 
 quantity, are capable of giving very easily recognisable 
 spectra ; and, accordingly, their presence has been discovered 
 in many substances in which it was formerly unsuspected. 
 Sodium salts are, indeed, almost universally diffused in 
 water, in the mineral constituents of the soil, in the tissues 
 of plants and animals, and in the dust suspended in the 
 atmosphere. 
 
 Spectroscopic analysis has also led to the discovery of 
 several elements previously unknown, and existing as un- 
 recognised impurities in various substances. These newly- 
 discovered elements are all metals. Their names are given 
 below : 
 
 NAME. ORIGINAL SOURCE. DISCOVERER. DATE. 
 
 Rubidium 1 Diirckheim Bunsen 1859 
 
 Caesium / Mineral Water 
 
 Thallium Seleniferous pyrites Crookes 1857 
 
 Indium Freiberg blende Reich & Richter 1863 
 
 Gallium Zinc blende Lecoq de 1875 
 from Pyrenees Boisbaudran 
 
 It will be seen by reference to the diagrams that the 
 spectra of the metals of the alkalis and of the alkaline 
 earths are comparatively simple, consisting, in most cases, 
 of a small number of lines, which are often widely separated 
 and easily recognised. Several other metallic elements, 
 under the same conditions, yield equally simple spectra : 
 that of thallium, for example, consisting of a single green 
 line, whilst the spectrum of indium exhibits one line in the 
 blue and another in the indigo. The spectra of the heavy 
 and less volatile metals are, however, in general, much more 
 
58 Chemical Philosophy. 
 
 complex ; the spectrum of iron, for instance, showing 
 upwards of four hundred and fifty lines, many of them 
 crowded together in the green. 
 
 When a compound, such as the chloride of a metal like 
 sodium, is heated in the Bunsen flame, it gives out light 
 which is very generally identical with the light obtained 
 from the incandescent vapour of the metal itself. Some- 
 times, however, this is not found to be the case, and the 
 spectra of certain compounds is different from that of either 
 of the constituents, taken separately, and is, in general, 
 more complex. 
 
 The accompanying diagram exhibits a comparison of 
 the spectra obtained by the introduction of solid calcium 
 
 chloride into the non-luminous gas-flame, and when the 
 electric spark is passed over the same compound. 
 
 The higher temperature in the second case causes a 
 decomposition of the compound, and the observed lines, 
 which are changed both in number and in refrangibility, are 
 attributed to the glowing vapour of the metal calcium itself. 
 
 The position of such lines is not altered at still higher 
 temperatures, though not unfrequently new lines make their 
 appearance. 
 
 An interesting method of applying the spectroscope to 
 the examination of solid bodies has been much employed 
 
Absorption Spectra. 59 
 
 in investigations into the nature of the rare earths which 
 contain yttrium and its allies. When the spark from a good 
 induction coil traverses a tube provided with a flat aluminium 
 pole at each end, and the air is then removed by the Sprengel 
 mercury-pump, the luminosity which the gas exhibits in the 
 early stages of the exhaustion gradually disappears, and a 
 dark space which at first surrounds the negative pole gra- 
 dually extends throughout the tube. When the luminous 
 cloud showing the presence of residual gas has almost dis- 
 appeared, the molecular discharge from the negative pole 
 begins to excite phosphorescence on the glass where it strikes 
 the side of the tube, and in any substance enclosed within 
 the tube. This occurs with the majority of substances only 
 when the pressure of the gas within has been reduced to 
 about one-millionth of an atmosphere. The emitted light 
 yields in most cases a faint continuous spectrum ; but in 
 some cases, especially the sulphates of the earths referred to 
 above, a spectrum is observed consisting of bright lines. 
 
 ABSORPTION SPECTRA. 
 
 When a transparent coloured medium, such as a piece 
 of glass or a coloured liquid, is brought into the path of a 
 ray of light before it enters the spectroscope, certain por- 
 tions of the spectrum disappear. The bands of darkness 
 which are thus produced are due to the interception of 
 certain portions of the light by the coloured glass or solu- 
 tion, or vapour, as the case may be. The resulting spec- 
 trum is called an absorption spectrum. It presents charac- 
 teristics which in many cases are as decided as those of 
 the emission spectrum, and the observation of absorption 
 spectra may occasionally be turned to practical account in 
 the recognition of the colouring matter of blood, of wines, 
 and many other substances, as well as of certain coloured 
 gases and vapours. 
 
 When the absorbing medium is the vapour of an element 
 
6o 
 
 CJumical Philosophy. 
 
 or of a compound which is volatile without decomposition, 
 the dark lines of absorption occupy the same position as 
 the bright lines in the spectrum of the light produced by 
 the ignition of the same vapour. Thus, if a beam of light 
 from a lamp is allowed to traverse a sufficiently thick stratum 
 of sodium vapour, two dark lines close together make their 
 appearance in the yellow ; and if such a spectrum is viewed 
 side by side with the ordinary spectrum of the monochromatic 
 sodium light, the position of the dark absorption band is 
 seen to coincide with that of the bright yellow lines. Such 
 phenomena are explained by the law that gases and vapours 
 are capable of absorbing and stopping the same rays of 
 light which they emit at higher temperatures, when in the 
 state of ignition or incandescence. 
 
 This law has been experimentally verified by the exami- 
 nation not only of the absorption spectrum of sodium, but 
 of other elements which give more complex spectra. The 
 facts thus established have been employed in the solution 
 of the very interesting problem presented by the light which 
 reaches us from the sun and other heavenly bodies. Sun- 
 light examined by a spectroscope exhibits a spectrum from 
 which none of the primary colours are absent, but which is 
 traversed by a very large number of fine black lines, many 
 of which were discovered so long ago as 1802 by Wollaston, 
 and examined in 1814 by a German optician, Fraunhofer. 
 They are usually known as Fraunhofer's lines. 
 
 C D 
 
 H 
 
 In the figure, the position of a few only of the most 
 prominent is indicated, and they are distinguished by the 
 letters used originally by Fraunhofer, 
 
 These dark spaces are now known to be absorption 
 bands. It has been found that their positions coincide with 
 
Heating and Chemical Effects of Radiant Energy. 6 1 
 
 the lines of the spectra of many terrestrial elements ; and 
 the coincidences which have been observed are so absolute 
 and so numerous as to lead to the inevitable conclusion that 
 they are produced by the vapours of such bodies existing in 
 the gaseous envelope of the sun. 
 
 The eye enables us to recognise only a part of the radia- 
 tion from a luminous object, namely, those rays only which 
 correspond to a rate of vibration lying within certain com- 
 paratively narrow limits, and which possess a particular 
 range of refrangibility. But accompanying these are others, 
 on the one hand of greater wave-length, and consequently 
 smaller refrangibility, which, in passing through a prism, 
 are less bent than the extreme red of the visible portion ; 
 and on the other hand there are rays of shorter wave-length 
 and greater refrangibility than those which are found at the 
 extreme visible portion of the violet. These rays, though 
 they give to the eye no impression of light, are capable of 
 producing both heating and chemical effects. They are 
 not, however, equally capable of producing such effects. 
 
 A sensitive thermometer exposed successively to different 
 portions of the visible spectrum will indicate that the heat- 
 ing effect is produced chiefly towards the red end, and even 
 that the maximum is attained amongst the rays which are 
 less refrangible than the extreme end of the visible red, and 
 the spectrum of which lies beyond in darkness. The yellow 
 and red luminous rays and the dark heating rays possess, 
 however, little or no chemical action so far as ordinary silver 
 salts are concerned, and accordingly photographers are 
 accustomed to manipulate their sensitive plates in rooms 
 illuminated by yellow or red light. The chemically active 
 rays are confined almost entirely to the violet end of the 
 spectrum ; the exact position of the most active being found, 
 however, to depend to a certain extent upon the nature of 
 the substance submitted to their influence. Thus, silver 
 salts are blackened, and hydrogen and chlorine are caused 
 to combine most rapidly, when exposed to that part of the 
 
62 
 
 Chemical Philosophy. 
 
 violet end situated between the lines G and H of the solar 
 spectrum, though more or less chemical activity is manifested 
 some distance beyond the visible violet on the one side and 
 as far as the middle of the green on the other ; all perceptible 
 action ceasing in the yellow or most luminous portion of the 
 visible spectrum. 
 
 These relations are exhibited by the three curves in the 
 following diagram : 
 
 The summit of each curve indicates the position of the 
 maxima of heating, luminous, and ordinary photographic 
 effects in the spectrum of sunlight, the vertical lines repre- 
 senting the positions of the chief dark lines in the solar 
 spectrum. 
 
 It would not be correct, however, to assume that rays in 
 other parts of the spectrum have no chemical activity, inas- 
 much as this property depends not only upon the colour 
 and refrangibility of the ray, but also upon the nature of the 
 substance upon which it falls. The chemical effects ob- 
 served in photographic plates are due to absorption of the 
 rays by the sensitive film; and when the composition of this 
 film is altered, or addition made of some appropriate colour- 
 ing matter, a photographic or chemical effect may be ob- 
 tained in any part of the spectrum from the region beyond 
 
Invisible Spectra. 63 
 
 the extreme visible violet to the dark region beyond the red. 
 These curves are represented as overlapping one another, 
 but it must not be supposed from this that there are three 
 distinct sets of rays in the spectrum. In all probability the 
 rays are of the same kind from end to end, but differ in 
 wave-length and rapidity of vibration. The least refrangible 
 rays are capable of producing the effects of heat, but are 
 not capable of exciting the sensations of vision ; whilst the 
 vibrations which communicate the sensation of violet to the 
 eye are also capable of effecting chemical change, though 
 their heating power is exceedingly small. 
 
 Modern researches having shown that invisible rays at 
 both extremities of the spectrum are capable of producing 
 a photographic image, when allowed to fall upon a plate 
 coated with suitable material, this discovery has rendered it 
 possible to observe the absorption-spectra of various trans- 
 parent media beyond the limits of the visible portion. The 
 most interesting results have been obtained with liquids 
 which are usually regarded as colourless, such as water, 
 alcohol, and various hydrocarbons. These liquids for the 
 most part absorb and stop certain rays both in the visible 
 portion of the spectrum and in the invisible regions at the 
 two extremities, and by the position of the dark bands or 
 lines thus observed they may be readily identified. The 
 only liquids which have at present been observed to give 
 no absorption bands either in the infra-red region or in the 
 visible spectrum are carbon tetrachloride and carbon disul- 
 phide. All those which contain hydrogen have been found 
 to give lines and bands of absorption. 
 
 So far as concerns the highly refrangible rays in the 
 ultra-violet, Professor Hartley concludes that though a 
 general absorption may be observed in the case of many 
 alcohols, acids, and hydrocarbons, it is only benzene and 
 compounds which contain six-carbon atoms united together 
 in the same manner as in benzene that give definite absorp- 
 tion bands. (See formula of benzene, p. 319.) 
 
64 Chemical Philosophy. 
 
 LUMINOSITY OF FLAME. 
 
 The cause of the luminosity of common candle and gas 
 flames has long been a subject of interest, and two chief 
 hypotheses have been framed with the object of explaining 
 it. Though neither of these hypotheses alone is capable of 
 furnishing a complete explanation of every case, yet, taken 
 together, they serve to account satisfactorily for the pheno- 
 mena which are usually observed. 
 
 According to the earlier hypothesis, proposed in 1817 
 by Sir H. Davy, the luminosity of flame is attributed to the 
 existence in the flame of particles of solid matter, which, 
 being heated to a high temperature by the burning gases, 
 emit light. These solid matters, supposed to exist in the 
 flame, are deposited in the form of soot when a cold body 
 is plunged into the flame. Such facts as the following were 
 adduced by Davy in support of his hypothesis. 
 
 The flame of hydrogen or of alcohol, burning in the 
 usual way, emits only a very feeble light ; but the intro- 
 duction of solid matter, such as powdered charcoal, oxide 
 of zinc, or dust of any kind, whether combustible or not, 
 serves to render such a flame luminous. 
 
 A bright flame is also produced by the combustion, in 
 air or oxygen, of substances which, like metallic zinc, form 
 solid products of combustion. On the other hand, sulphur, 
 hydrogen, and carbonic oxide, which, in burning, yield 
 entirely gaseous products, give out light very sparingly. 
 
 The spectroscope has also shown us that ordinary lumi- 
 nous flames yield a continuous spectrum of the same cha- 
 racter as that usually produced by incandescent solids. 
 
 The production of solid particles in a candle-flame, or 
 in the flame of coal-gas, or other hydrocarbon, is supposed 
 to be due to the selective power of the oxygen of the air, 
 in virtue of which it unites preferably with the hydrogen, 
 
Luminosity of Flame. 65 
 
 leaving a part of the carbon in a solid form to be consumed 
 as it reaches the higher parts of the flame. 
 
 A candle flame burning steadily exhibits the form of a 
 long cone with the apex pointing upwards. Other flames 
 taper upwards in the same way. This form is, of course, 
 due to the strong upward current produced in the air which 
 immediately surrounds the flame, in consequence of the 
 heat. A close examination of the flame of a candle will 
 show that it consists of several parts At the base is a 
 faintly luminous stratum of a blue colour. Higher up the 
 luminosity increases, the most intense light proceeding from 
 those parts which are near the middle ; whilst the whole 
 structure is surrounded by a transparent and almost invisible, 
 but very hot, envelope of ignited gas. It is easy to show 
 by many simple experiments that such a flame is only a 
 shell of ignited gas, the process of combustion occurring 
 only on the outside, the interior being filled with compara- 
 tively cool gas or vapour. Thus, if a sheet of paper is held 
 for a moment horizontally across the flame it will receive a 
 deposit of soot in the form of a ring. A slip of wood or a 
 wire passed through the flame becomes ignited only at the 
 two points where it cuts the outer portions of the flame, the 
 part in contact with the interior remaining dark. 
 
 The luminosity of such a flame is diminished if a suffi- 
 cient quantity of air or oxygen is thrown into it, and at the 
 same time it loses the power of depositing soot upon any 
 cold object held in it. The flames of the blowpipe and the 
 Bunsen burner are produced in this way, and present this 
 character. 
 
 Some experiments made a few years ago by Dr. Frank- 
 land indicate that the ignition of solid particles of sooty 
 matters in hydrocarbon flames may occasionally be the 
 cause of the light emitted by such flames, but that in some 
 cases at least the effect is not wholly attributable to this cir- 
 cumstance. Dr. Frankland has observed that when hydrogen 
 is burnt in oxygen under great pressure the light of the 
 
66 Chemical Philosophy. 
 
 flame, usually so pale, is increased to such an extent as to 
 be capable of illuminating a page of print so that it can be 
 read at some distance. 
 
 The spectrum of hydrogen burning under pressure ex- 
 hibits the three bands characteristic of hydrogen, but much 
 broader and more or less nebulous at the edges, so that an 
 approach to a continuous spectrum similar to that obtained 
 from solid bodies is the result. A similar effect is pro- 
 duced by burning carbonic oxide. So that the argument 
 founded upon the nature of the spectrum of an ordinary 
 luminous flame has less weight than might otherwise be 
 supposed. 
 
 Moreover, in burning a series of substances, all of which 
 yield volatile products of combustion, it is found that many 
 of them are capable of emitting a very vivid light, quite 
 equal in brilliancy to the light produced by the ignition 
 of many substances which are solid and not capable of 
 vapourisation. Thus, the combustion of arsenic in oxygen 
 is attended by the emission of a very brilliant white light, 
 although all the substances present the arsenic, the oxygen, 
 and the arsenious oxide which is formed are, at the tem- 
 perature of ignition, entirely in a state of vapour. 
 
 In such cases, therefore, the light cannot be attributed 
 to glowing solid matter. It has been observed that the 
 brilliancy of the light emitted by the combustion of such 
 substances is nearly proportional to the density of the ignited 
 vapours existing in the flame, provided that in each case the 
 temperature is sufficiently high. 
 
 In all probability, then, the luminosity of burning gas or 
 tallow is due to the ignition of vaporous, but very dense, 
 hydrocarbons, and is not to be ascribed to the presence of 
 solid particles of carbon. The fact that ordinary flames 
 are transpaient is also difficult to reconcile with the latter 
 hypothesis. 
 
Exercises on Section I. 67 
 
 EXERCISES ON SECTION I. 
 
 1. Water is shaken up with a large volume of oxygen gas under a 
 constant pressure of 765 mm. What volume of the gas will be con- 
 tained in 10 c.c. of the solution? 
 
 2. Water is exposed to an atmosphere consisting of 21 vols. of 
 oxygen, with 79 volumes of nitrogen. 
 
 Temp. o ; pressure, 760 mm. 
 Coeffs. of sol., N ^ -02, O = -04. 
 
 Calculate (a) the total volume of gas dissolved by 52-5 c.c. of 
 water, and (b] the percentage composition of the gas. 
 
 3. Soda-water is charged under a pressure of 2 '3 atmospheres. 
 Calculate the volume of carbonic anhydride contained in 300 cubic 
 centimetres of such water. 
 
 An atmosphere = 760 mm. barom. 
 
 Coeff. of sol. for carbonic anhydride, 1 7967. 
 
 4. Sulphur dioxide is passed into water as long as it is absorbed. 
 If the barometer stands at 745 mm., calculate the volume of gas con- 
 tained in half a litre of the solution. 
 
 Coeff. for sulphur dioxide, 79*789. 
 
 5. Water is shaken up with its own bulk of a mixture of I volume 
 of oxygen with 3 vols. of nitrogen. Supposing the temp, and pressure 
 to remain normal and constant throughout the experiment, calculate the 
 composition of the residual air. 
 
 Coeff. of oxygen, '04114. 
 Coeff. of nitrogen, -02035. 
 
 Diffusion of Gases. 
 
 6. The specific gravity of chlorine is 35-5 (H = i). Compare its 
 velocity of diffusion with that of hydrogen. 
 
 7. Specific gravity of ozone, 24 ; of carbonic anhydride, 22. Com- 
 pare their velocities of diffusion with each other and with that of H 
 (sp. gr. I). 
 
 8. The rate of diffusion of a gas is observed to be '81 when that of 
 air is I. Find its density. 
 
 9. Oxygen and hydrogen are separated by a porous plale, and 3-83 
 cubic centimetres of hydrogen pass through the plate in a second. 
 
 F 2 
 
68 Chemical Philosophy. 
 
 What volume of oxygen passes during the same time in the opposite 
 direction ? 
 
 10. In the last question suppose the original volume of the oxygen 
 to have been 20 c.c. : what will be the composition of the mixture formed 
 in its place after three seconds, assuming the apparatus so arranged that 
 no change of pressure occurs ? 
 
 Corrections of Gas-volumes for Changes of Pressure and Temperature. 
 
 11. loo c.c. of air when bar. = 750 mm. Find the volume when 
 bar. = 790 mm. 
 
 12. 250 c.c, of air when bar. = 765 mm. Find the volume when 
 bar. = 745 mm. 
 
 13. What pressure in atmospheres would be required to make the 
 density of hydrogen (sp. gr. "0693) equal to that of air ? 
 
 14. Calculate the atmospheric pressure per square centimetre when 
 the barometer stands at 760 mm. 
 
 Weight of I c.c. of mercury, 13*596 grams. 
 
 15. Find the atmospheric pressure per square decimetre when the 
 barometer stands at 750 mm. 
 
 16. What change of atmospheric pressure will be denoted by a 
 change of 12 mm. in the barometric column ? 
 
 17. The weight of one litre of hydrogen at o and 760 mm. is 
 0896 gram or I crith. Calculate the weight of one litre of hydrogen 
 measured off under a pressure of 1400 mm. 
 
 18. Find the weight of I litre of nitrogen (sp. gr. 14) ; of 10 litres 
 of carbonic anhydride (sp. gr. 22) ; of 250 c.c. of oxygen (sp. gr. 16). 
 
 19. A mass of air at o measures 1 00 c.c. What volume will it 
 occupy at 20 ; at 1 5 -5 ; at 100 ! 
 
 20. A certain- quantity of air is measured at 75. What volume 
 will it have at o ? 
 
 21. 1000 c.c. of a gas at I2'5. What volume at 75 ? 
 
 22. 500 c.c. of a gas at 10. What volume at 40 ? 
 
 23. 300 c.c. of a gas at 25. What will its volume be at 10 ? 
 
 24. 75 c.c. of nitrogen measured at 50. What volume at 35? 
 
 25. 1500 c.c. of hydrogen measured at 20. At what temperature 
 will it measure 1000 c.c.? 
 
 26. Five degrees centigrade correspond with nine degrees on the 
 Fahrenheit scale. Find the coefficient of expansion of gases for i F. 
 
 27. 150 c.c. of nitrogen are measured at 10, and under a pressure 
 
Exercises on Section I. 69 
 
 of 500 mm. of mercury. What will the volume become at 1 6 -4 when 
 the pressure is 540 mm.? \ y. >"""* 7 V 
 
 28. A quantity of nitrogen confined in a tube standing over mercury 
 in a mercurial trough measures 75-5 c.c.; temp. 15; bar. 742 mm.; 
 surface of mercury inside the tube above surface of mercury in the 
 trough 122 mm. Find the volume which the gas would occupy at 
 normal temperature and pressure. 
 
 29. i ooo cubic feet of gas are put into a balloon of 1250 cubic feet 
 capacity ; temp. 18 ; bar. 765 mm. After ascending a certain height 
 it is found to be fully distended. What is the atmospheric pressure, 
 temperature being 8 ? 
 
 30. A certain balloon is just capable of holding 10 grams of hydrogen 
 under standard conditions : what is its capacity ? How much larger 
 must it be made if it is required to sustain a diminished atmospheric 
 pressure equal to 650 mm. bar. ? 
 
7O Chemical Philosophy. 
 
 SECTION II. 
 
 CHAPTER IX. 
 
 ELEMENTS AND COMPOUNDS. 
 
 WHEN water is exposed to a very high temperature or 
 made the vehicle of an electric current, it disappears and is 
 replaced by an equal weight of a mixture of two gases, 
 hydrogen and oxygen. These two gases will, under certain 
 conditions, again give rise to water and to exactly the same 
 amount of water as at first. 
 
 Water then is said to be composed of oxygen and 
 hydrogen. It is worth noting, however, that, strictly speak- 
 ing, this can only mean that in proportion as the water is 
 destroyed or ceases to exist, the gases make their appearance, 
 and vice versa, for in water we have no resemblance to 
 hydrogen or oxygen, neither can we detect either of those 
 bodies in water except by this process of so-called de- 
 composition. * 
 
 Now, if the hydrogen thus obtained from water is sub- 
 mitted to a repetition of the same kind of treatment or to 
 any other that may suggest itself, it refuses utterly to yield 
 up anything that is not hydrogen. In other words, it cannot 
 be decomposed. We find then that certain bodies, 2 such 
 
 1 ' Cavendish and Watt both discovered the composition of water. 
 Cavendish established the facts ; Watt, the idea. Cavendish says, 
 " From inflammable air and dephlogisticated air water is produced." 
 Watt says, "Water consists or is composed of inflammable air and 
 dephlogisdcated air." Between these forms of expression there is a 
 wide distinction.' Liebigs Letters on Chemistry, p. 58. 
 
 a It must be understood that bodies of definite characters, and not 
 mere mixtures, are here referred to. (See Chap, XII.) 
 
Elements and Compounds. 71 
 
 as water, may be resolved into two or more different kinds 
 of matter, and these are called compounds ; whilst others like 
 hydrogen cannot be split up in this manner by any means 
 with which we are at present acquainted, and are regarded 
 as elements. 
 
 It must be understood that the word ' element ' is only 
 provisional, and is merely intended to imply that in the 
 present state of knowledge the bodies so called have resisted 
 all attempts to decompose them. 
 
 Concerning the primal elements of nature nothing what- 
 ever is known. Some speculations concerning the constitu- 
 tion of the chemical ' elements ' will be referred to later on, 
 but it will be sufficient in this place to indicate the general 
 nature of the investigations that have been made in regard 
 to one or two of them. 
 
 Chlorine was discovered in 1774 by Scheele, who, in the 
 language of the then prevalent theory, called it ' dephlogis- 
 ticated muriatic acid.' This was equivalent to saying that it 
 was muriatic acid deprived of its inflammable principle, 
 hydrogen. After the discovery of the fact that many of the 
 common acids contain oxygen, Berthollet, acting under the 
 belief that it contained oxygen, gave it the name ' oxymuriatic 
 acid.' Davy proved, in 1810, that it contained neither hy- 
 drogen nor oxygen, and gave it the name by which it has 
 since been known and ranked with the rest of the elements. 
 
 Davy's chief experiments were as follows : 
 
 The dry gas mixed with dry hydrogen unites with it with- 
 out contraction and without the production of any water. 
 Charcoal, previously freed from hydrogen and moisture, 
 undergoes no change when intensely ignited in the dry gas 
 by the voltaic current even for several hours, and no oxide 
 of carbon is formed. Potassium, sodium, tin, copper, and 
 other metals when heated in it combine with the whole of 
 the gas, leaving no gaseous residue and forming no oxide of 
 the metal. Phosphorus and sulphur, both highly oxidisable 
 substances, unite with chlorine but set free from it no 
 
72 Chemical Philosophy. 
 
 muriatic acid. Chlorine mixed with dry ammonia produces 
 only nitrogen and dry sal-ammoniac, and no water is formed. 
 Electric sparks passed for a long time through chlorine gas 
 produce no change in it. When chlorine acts upon a 
 heated metallic oxide the amount of oxygen expelled is 
 equal to that which is contained in the oxide and no more. 
 So much for direct experimental evidence. The discovery 
 of iodine, in 1811, and of bromine, in 1826, confirmed the 
 position of chlorine among the elements, for if oxygen be a 
 constituent of one of these bodies it must be present in 
 them all. Further knowledge since obtained, the accurate 
 determination of their atomic weights and the place they 
 occupy in the system of classification which is based upon 
 the numerical values of the atomic weights, establishes the 
 claim of these three substances to be regarded as a natural 
 group of elements. (See Chap. XXIII.) 
 
 The case of the fixed alkalies, potash and soda, is the 
 converse of that of chlorine. Originally supposed to be 
 elementary in consequence of the failure of all attempts to 
 extract from either of them more than one form of matter, 
 they had been suspected by Lavoisier to contain nitrogen, 
 probably by reason of a certain resemblance to the volatile 
 alkali ammonia, which was known to contain both hy- 
 drogen and nitrogen. By some of the same school it had 
 been guessed that the alkalies and alkaline earths were 
 compounds of metals with oxygen, in consequence of the 
 resemblance which some of them, such as lime, present 
 to oxides, like litharge, the composition of which was 
 known. In 1807 it was known that the voltaic current 
 would decompose water into oxygen and hydrogen, and 
 that solutions of various salts were resolved into acid, which 
 collected at the positive pole, and alkali, which always 
 passed to the negative, also that many of the metals and their 
 oxides might be separated from their compounds by the 
 same agency. With these facts in view Davy exposed 
 caustic potash to the action of a powerful electric current 
 
Elements and Compounds. 73 
 
 and obtained oxygen and potassium. Caustic soda was by 
 the same means shown to contain sodium. These new 
 metals were at first supposed to be compounds of hydrogen 
 with potash and soda respectively, but in that case they 
 ought, when burnt in oxygen, to produce the hydrates. 
 That this is not the case was shown by Davy. 
 
 As to potassium and sodium their elemental nature 
 seems almost above suspicion, for there is no class of sub- 
 stances possessing so many characters in common as the 
 metals, and if it could be shown that any one of them con- 
 tained, for example, hydrogen, the same constituent might 
 be expected to occur in all. The results of spectroscopic 
 analysis, however, seem to place this question beyond doubt, 
 for the spectrum of each metal is independent, and shows 
 no sign of containing an element in common with any of 
 the others. 
 
 Other examples might be mentioned of substances long 
 regarded as chemical elements ultimately showing their true 
 character as compounds under the influence of some new 
 agent or some improved mode of operating ; thus, the 
 substance regarded as metallic vanadium by Berzelius turned 
 out to be an oxide of vanadium when examined long after- 
 wards by Roscoe. 
 
 In the case of oxygen it is found that, by the action of 
 electricity and otherwise, it may be converted into another 
 gas, ozone, possessing remarkable characters quite distinct 
 from those of oxygen. Nevertheless, oxygen ranks as an 
 element because it yields in this way only one new body at 
 a time, which by mere application of heat recovers its original 
 properties, and that without loss or gain in weight. It must, 
 therefore, be assumed that the altered properties exhibited 
 under these circumstances result from a temporary re- 
 arrangement of its constituent particles. When, as in this 
 case, elementary matter, stuff, or substance is capable of 
 making its appearance in the form of two or more bodies 
 having different properties, these are said to be allotropic 
 
74 
 
 Chemical Philosophy. 
 
 modifications of the element, and the phenomenon is spoken 
 of as allotropy. (See 'Isomerism,' Chap. XXL) 
 
 About seventy elements are known at the present time, 
 but it is not improbable that a few new substances may be 
 hereafter added to this number. It is not, however, very 
 likely that any hitherto unknown elements will be found to 
 occur in any considerable quantity among the constituents 
 of the earth's crust, for every substance within reach of man 
 has already been subjected to a very close scrutiny by 
 chemists. 
 
 It will be seen by reference to the following table that 
 the materials which compose the solid earth, so far as we 
 know it the ocean, the atmosphere, and the bodies of the 
 living beings which inhabit it are made up of a few of these 
 elements, the rest occurring in much smaller quantity, in 
 some cases discoverable only by specially delicate methods. 
 
 Water consists of . 
 Air consists chiefly of . 
 Silica 
 
 Solid earth con-7 
 sists chiefly of 
 
 Limestone 
 
 Various silicates 
 forming crystal- 
 line rocks 
 beds of clay 
 
 or 
 
 Plants consist chiefly of . 
 
 j Hydrogen. 
 (Oxygen. 
 
 (Oxygen. 
 (Nitrogen. 
 
 ( Silicon. 
 I Oxygen. 
 
 Calcium. 
 
 Magnesium. 
 
 Carbon. 
 
 Oxygen. 
 
 Silicon. 
 
 Oxygen. 
 
 Aluminium. 
 
 Iron. 
 
 Calcium. 
 
 Potassium. 
 
 f Carbon. 
 Hydrogen. 
 Oxygen. 
 
Symbols. 75 
 
 Carbon. 
 
 / 
 
 Soft Parts 
 
 Animals consist 
 
 Hydrogen. 
 Oxygen. 
 
 Nitrogen. 
 
 chiefly of Calcium 
 
 tBoneandShen= 
 
 I Carbonate. 
 
 Chemists have found it convenient to adopt a system of 
 symbols and formulae, whereby to represent the elements 
 and those compounds the composition of which is known. 
 
 Thus, to each of the elements is assigned a symbol 
 formed generally of the initial letter of the Latin name of 
 the element. For example, sulphur is represented by the 
 symbol S ; selenion, Se ; silicon, Si ; strontium, Sr ; but 
 silver (argentuni) is Ag, and sodium (natron) has the 
 symbol, Na. 
 
 But chemical symbols are not merely abbreviations 
 contrived, like shorthand characters, for the purpose of 
 saving trouble in writing the names. Each symbol repre- 
 sents one atom of the element for which it stands, and hence 
 expresses a definite weight and volume of the element, 
 which are identical with the proportions by weight and 
 volume in which it enters into chemical combination. 
 
 The weight of an atom of hydrogen is less than the 
 weight of an atom of any other element. It is, therefore, 
 convenient to consider the value of the symbol of hydrogen 
 as unity and the values of all other symbols greater than 
 unity. But since nothing is known as to the absolute 
 weight of an atom of hydrogen or of any other element, it 
 should be borne in mind that the atomic weights are in 
 reality ratios, or fractions whose denominator is i, although 
 they are always written in the form of whole numbers. 
 
 In the following list are given the names of all the known 
 elements, together with their symbols and atomic weights, 
 and the student is recommended at once to commit to 
 memory those which are printed in capital letters, leaving 
 
76 Chemical Philosophy. 
 
 the rest to be learnt gradually as occasion may require. In 
 a later chapter an account will be given of the methods by 
 which the numerical values of the atomic weights are deter- 
 mined. 
 
 TABLE OF ATOMIC WEIGHTS. 
 
 Name of Element 
 
 Symbol 
 
 Calculated from 
 the most accurate 
 Experiments 
 (H = i) 
 
 Approximate 
 Values for 
 Common Use, 
 O=i6 
 
 ALUMINIUM 
 
 Al 
 
 27-009 
 
 27 
 
 Antimony . 
 
 Sb (Stibium) 
 
 II9-955 
 
 120 
 
 Arsenic 
 
 As ... 
 
 74'9^ 
 
 75 
 
 Barium 
 
 Ba 
 
 136763 
 
 137 
 
 Beryllium 
 
 Be ... 
 
 9-085 
 
 9-1 
 
 Bismuth 
 
 Bi 
 
 207-523 
 
 208 
 
 Boron . 
 
 B 
 
 10-941 
 
 ii 
 
 BROMINE. 
 
 Br 
 
 79-768 
 
 80 
 
 Cadmium 
 
 Cd . 
 
 III-835 
 
 112 
 
 Caesium 
 
 Cs 
 
 I32-583 
 
 133 
 
 CALCIUM . 
 
 Ca . 
 
 39-990 
 
 40 
 
 CARBON . 
 
 C 
 
 11-9736 
 
 12 
 
 Cerium 
 
 Ce . 
 
 140-424 
 
 141 
 
 CHLORINE 
 
 Cl 
 
 35*370 
 
 35-5 
 
 Chromium . 
 
 Ci 
 
 52-009 
 
 52-1 
 
 Cobalt 
 
 To ... 
 
 58-887 
 
 CO 
 
 COPPER . 
 
 Cu (Cuprum) 
 
 j u ""/ 
 63-173 
 
 J-7 
 
 63-3 
 
 Didymium . 
 
 Di . . 
 
 
 145? 
 
 Erbium 
 
 E 
 
 165-891 
 
 166 
 
 FLUORINE 
 
 F 
 
 18-984 
 
 19 
 
 Gallium 
 
 Ga . 
 
 68-854 
 
 69 
 
 Germanium . 
 
 Ge . 
 
 72-32 
 
 72 
 
 Gold . 
 
 Au ( Aurum) . 
 
 196-850 
 
 197 
 
 Holmium 
 
 Ho . 
 
 ? 
 
 v 
 
 HYDROGEN 
 
 H 
 
 I -0000 
 
 I 
 
 Indium 
 
 In ... 
 
 113-398 
 
 114 
 
 IODINE . 
 
 I ... 
 
 127-557 
 
 127 
 
 Indium 
 
 Ir 
 
 192-651 
 
 193 
 
 IRON. 
 
 Fe (Ferrum) . 
 
 55-913 
 
 56 
 
 Lanthanum . 
 
 La ... 
 
 138-526 
 
 139 
 
 LEAD 
 
 Pb (Plumbum) 
 
 206-471 
 
 207 
 
 Lithium 
 
 Li ... 
 
 7-0073 
 
 7 
 
 MAGNESIUM 
 
 Mg . 
 
 23-959 
 
 24 
 
 Manganese . 
 
 Mn . 
 
 53-906 
 
 54 
 
 MERCURY 
 
 Hg (Hydrargyrum) 
 
 199-712 
 
 200 
 
 Molybdenum 
 
 Mo ... 
 
 95-527 
 
 957 
 
 Nickel. 
 
 Ni . 
 
 57-928 
 
 58 
 
Symbols and Atomic Weights. 
 
 77 
 
 TABLE OF ATOMIC WEIGHTS Continued. 
 
 Name of Element 
 
 Symbol 
 
 Calculated from 
 the most accurate 
 Experiments 
 (H = i) 
 
 Approximate 
 Values for 
 Common Use, 
 O=i6 
 
 Niobium 
 
 Nb . 
 
 93-812 
 
 94 
 
 NITROGEN 
 
 N 
 
 I4-O2IO 
 
 14 
 
 Osmium 
 
 Os . 
 
 198-494 
 
 199 
 
 OXYGEN . 
 
 
 
 15-9633 
 
 16 
 
 Palladium . 
 
 Pd 
 
 105-737 
 
 106 
 
 PHOSPHORUS 
 
 P 
 
 30-958 
 
 3i 
 
 Platinum 
 
 Pt 
 
 194-415 
 
 195 
 
 POTASSIUM 
 
 K (Kalium) . 
 
 39-019 
 
 39-1 
 
 Rhodium 
 
 Ro 
 
 104-055 
 
 104 
 
 Rubidium . 
 
 Rb . 
 
 85-251 
 
 85-5 
 
 Ruthenium . 
 
 Ru 
 
 104-217 
 
 104 
 
 Samarium . 
 
 Sm . , 
 
 ? 
 
 ? 
 
 Scandium 
 
 Sc 
 
 43-980 
 
 44 
 
 Selenion 
 
 Se . 
 
 78-797 
 
 79 
 
 SILICON . 
 
 Si 
 
 28-195 
 
 28 
 
 SILVER . 
 
 Ag (Argentum) 
 
 107-675 
 
 108 
 
 SODIUM . 
 
 Na (Natrium) 
 
 22-998 
 
 23 
 
 Strontium . 
 
 Sr . 
 
 87-374 
 
 87-6 
 
 SULPHUR 
 
 S 
 
 31-984 
 
 3 2 
 
 Tantalum 
 
 Ta 
 
 182-144 
 
 182 
 
 Tellurium . 
 
 Te . 
 
 127-960 
 
 126? 
 
 Thallium . 
 
 Tl 
 
 203-715 
 
 204 
 
 Thorium 
 
 Th 
 
 233-414 
 
 234 
 
 Thulium 
 
 Tm . 
 
 9 
 
 v 
 
 Tin . 
 
 Sn (Stannurn) 
 
 117-698 
 
 118 
 
 Titanium 
 
 Ti 
 
 48-013 
 
 48 
 
 Tungsten 
 
 W (Wolfram) 
 
 183-610 
 
 184 
 
 Uranium 
 
 U 
 
 238-482 
 
 239 
 
 Vanadium . 
 
 V 
 
 51-256 
 
 5!"4 
 
 Ytterbium . 
 
 Yt 
 
 172-761 
 
 173 
 
 Yttrium 
 
 Y 
 
 89-816 
 
 90 
 
 ZINC . 
 
 Zn 
 
 64-9045 
 
 65 
 
 Zirconium . 
 
 Zr 
 
 89-367 
 
 89-6 
 
 In order to indicate combination between two elements 
 their symbols are placed side by side, thus, HC1. When a 
 molecule of a compound contains more than one atom of 
 each or either of its constituents, the number of atoms is 
 indicated by a small figure placed below the line. 
 
78 Chemical Philosophy. 
 
 Thus the formula for water, OH 2 , represents one atom of 
 oxygen united with two atoms of hydrogen, and H 3 PO 4 
 means that three atoms of hydrogen, one atom of phos- 
 phorus, and four atoms of oxygen are bound together in 
 phosphoric acid. 
 
 The formula, taken as a whole, is generally assumed- to 
 represent a molecule of the compound, though it will here- 
 after (Chapter XV.) be explained that much uncertainty exists 
 in reference to all substances except such as are vaporis- 
 able without chemical decomposition. The relative weight 
 of the molecule is easily found by adding together the 
 weights represented by the several symbols of which it is 
 made up. This weight is called the molecular weight. In 
 order to express two or more molecules a figure is placed 
 at the beginning of the formula, and must be understood 
 to multiply every symbol that follows it. For example, 
 2OH 2 represents two molecules of water, each consisting of 
 two atoms of hydrogen with one atom of oxygen ; in all, 
 four atoms of hydrogen and two atoms of oxygen. Occa- 
 sionally brackets have to be introduced when some group 
 of symbols occurs more than once. Aluminium sulphate, 
 for example, has the formula A1 2 (SO 4 ) 3 , which is thus 
 written for the sake of assimilating the appearance of the 
 formula to that of other sulphates, such as H 2 SO 4 , BaSO 4 , 
 FeSO 4 , &c. Resemblance between them would be less 
 apparent if it were expressed as A1 2 S 3 O 12 . 
 
 Care and a little practice are all that is necessary to 
 avoid confusion in the use of formulae, arid the student is 
 therefore recommended to work out conscientiously all the 
 examples given at the end of the section. 
 
Laws of Chemical Combination. 79 
 
 CHAPTER X. 
 
 LAWS OF CHEMICAL COMBINATION. 
 
 THE extraction of the metals from their ores, the manufac- 
 ture of alkalies, soap, glass, dyes, and a variety of other 
 useful applications of practical chemistry, were known to 
 man in a more or less practical form from very early times. 
 The alchemists extended the art of chemistry by the dis- 
 covery of the processes for producing many acids and salts, 
 arid by the invention of much useful apparatus. The dis- 
 covery of oxygen by Priestley, of chlorine by Scheele, the 
 proof of the composition of water supplied by Cavendish, 
 and the overthrow of the phlogistic theory by Lavoisier's 
 explanation of combustion, were all great strides in advance 
 which followed one another in rapid succession. But the 
 foundation of chemistry as an exact science was only laid 
 towards the end of the last century, when the balance 
 began to be used systematically in all chemical investiga- 
 tions. Exact determinations of the relative weights of 
 bodies engaged in various chemical reactions were necessary 
 for the establishment of those laws upon which chemical 
 ideas of the present day are founded. If we start by admit- 
 ting the molecular constitution of matter, it becomes un- 
 necessary to make any formal statement of the laws which 
 have been found to regulate the distribution of weight or 
 quantity of matter involved in chemical actions. If we 
 assume, for example, that the element oxygen is made up of 
 molecules, all having the same properties, and each composed 
 of two atoms each having a weight sixteen times that of the 
 hydrogen atom, it is obvious that only a definite number of 
 those atoms can take part in forming a chemical compound. 
 And if, as we believe, these atoms are indestructible by 
 chemical action, no fractional part of an atom can enter into 
 
8o Chemical Philosophy. 
 
 such process ; and if the number of atoms thus employed 
 be represented as n, the weight of the substance will be i6n. 
 But since the molecular theory was not always employed 
 by chemists, and even at the present day is not adopted 
 universally, it is desirable still to indicate the facts which 
 have been observed in connection with this question. The 
 four general statements or laws may be expressed as follows : 
 
 LAW OF DEFINITE PROPORTIONS. 
 
 The proportions in which bodies unite together chemi- 
 cally are definite and constant. In other words, a given 
 chemical compound always consists of the same elements 
 united in the same proportions. 
 
 In order to form water, for example, union between 
 hydrogen and oxygen occurs exactly in the proportion of 
 two measures of the former to one measure of the latter. 
 This corresponds with two parts by weight of hydrogen to 
 sixteen parts by weight of oxygen, since oxygen is sixteen 
 times heavier than hydrogen. The employment of any 
 larger quantity of either element would only result in the 
 excess being left uncombined. 
 
 LAW OF MULTIPLE PROPORTIONS. 
 
 When two substances unite together in several propor- 
 tions, then if one of them is taken as the unit, the quantity 
 of the other substance in the different compounds varies in 
 a simple manner. Oxygen combines with carbon, forming 
 two oxides of carbon. In the one, three parts by weight of 
 carbon are united to four parts of oxygen, whilst in the other 
 three parts of carbon combine with twice this quantity or 
 eight parts of oxygen. A great number of equally simple 
 cases might be cited, and, doubtless, it was the study of such 
 cases which in the first instance led to the enunciation of 
 the law. In perhaps a still greater number of instances, 
 however, it is by no means easy to trace its application. 
 Among the numerous compounds of carbon with hydrogen, 
 
Multiple and Reciprocal Proportions. 8 1 
 
 for example, are found such relations as the following, which 
 represent the composition of the series of paraffins : 
 
 Carbon. Hydrogen. 
 
 Methane ... 3 P ar ts with i part 
 
 Ethane .... 4 i 
 
 Propane ... 9 2 
 
 Butane. .24 5 
 
 Pentane ... 5 i 
 
 Hexane . .36 7 
 
 Heptane . . .21 4 
 
 Octane 16 3 
 
 Nonane ... 27 5 
 
 Decane. , , .60 n 
 
 &c. &c. &c. 
 
 LAW OF RECIPROCAL PROPORTIONS. 
 
 The weights of two different elements, A and B, which 
 combine with a third, C, represent the proportions in which 
 they will themselves unite together if union between them 
 is possible, or they bear some simple relation to those pro- 
 portions. Thus, 35^ parts of chlorine and 80 parts of 
 bromine combine with 23 parts of sodium. Then, accord- 
 ing to the law, when chlorine combines with bromine, 
 35^ parts of the former are required for every 80 parts of 
 the latter. This may be rendered graphically somewhat in 
 this manner : 
 
 P 35'5 
 
 Chlorine. 
 
82 Chemical Philosophy. 
 
 In other words, the combining weight of a body may always 
 be represented as a multiple of the same number whatever 
 state of combination it enters into, or m n is the combining 
 weight, where m is a number peculiar to the body and n is 
 some integer. 
 
 LAW OF VOLUMES. 
 
 When gases combine together they do so in equal 
 volumes, or in volumes which have some simple relation one 
 to another, as i to 2, i to 3, 2 to 3, and so on. For 
 example, hydrochloric acid is formed by the union of 
 i volume of hydrogen with 
 
 1 volume of chlorine ; 
 
 water is formed by the union of 
 
 2 volumes of hydrogen with 
 i volume of oxygen ; 
 
 ammonia by the combination of 
 
 3 volumes of hydrogen with 
 
 1 volume of nitrogen ; 
 
 nitrous anhydride by the combination of 
 
 2 volumes of nitrogen with 
 
 3 volumes of oxygen, &c. 
 
 These facts are recorded in the formulae 
 
 HC1 H 2 H 3 N N 2 3 
 
 by which these compounds are commonly represented. 
 
 It is important also to remember that whatever the 
 volume of the vaporous elements before combination, the 
 bulk of the resulting compound, measured in the gaseous 
 state under the same conditions of temperature and pres- 
 sure, is generally less than the total bulk of the constituent 
 gases measured separately. 
 
 In a very few cases the volume occupied by the united 
 substances is the same whether they are combined chemically 
 or only mixed together. 
 
Gay-Lussac's Lazv of Volumes. 83 
 
 Thus: 
 
 i volume of hydrogen with 
 
 1 volume of chlorine 
 
 mixed together would of course form only two volumes of 
 the mixture, but when united they also produce 
 
 2 volumes of hydrochloric acid gas. 
 
 On the other hand, 
 
 1 volume of oxygen with 
 
 2 volumes of hydrogen 
 
 contract into 
 
 2 volumes of water vapour. 
 
 This will be again referred to in connection with the 
 methods employed for the determination of molecular 
 weights, and until those subjects have been discussed the 
 student must be content to take upon trust all statements 
 concerning molecular volumes, weights, and formulae. 
 
 This law was first enunciated by Gay-Lussac very soon 
 after the laws of reciprocal and multiple proportions by 
 weight had been established by Dalton. We now see the 
 connection between them and the dependence of the one 
 upon the other, though this relation, from want of accurate 
 knowledge, was not at first admitted by Dalton and other 
 chemists. Originally a fact, depending upon experimental 
 evidence only, this statement of Gay-Lussac's now stands in 
 the more important position of a logical deduction from 
 another law, the law of Avogadro (Chap. VII. p. 51). 
 
 The results of analysis before Dalton's time were repre- 
 sented by numbers which expressed the proportion of each 
 constituent in 100 parts of the compound. For example, 
 in 100 parts of red oxide of copper there are 
 
 Copper .... 88-8 
 Oxygen . . . . 11*2 
 
 100*0 
 
 G 2 
 
$4 Chemical Philosophy. 
 
 In black oxide of copper : 
 
 Copper .... 79'Sy 
 Oxygen .... 20-13 
 
 lOO'OO 
 
 In carbonic oxide : 
 
 Carbon . . . . 42-85 
 Oxygen .... 57-14 
 
 lOO'OO 
 
 In carbonic anhydride : 
 
 Carbon .... 
 Oxygen .... 
 
 TOO'OO 
 
 Dalton discovered reciprocal and multiple proportions 
 by stripping from them the disguise in which this mode of 
 representing composition enveloped them. For by taking 
 the proportion of some one of the elements in a series of 
 similar compounds as unity, and ascertaining by calcula- 
 tion the proportions which the others bear to it, it is easy to 
 show that these proportions are simple multiples one of 
 another. 
 
 In the case of the two oxides of carbon, the composition 
 of which has just been stated, if we calculate the ratio of the 
 oxygen to the carbon in both of them, we find the propor- 
 tion in the second double that in the first. Thus, taking 
 the carbon as unity 
 
 Carbon. Oxygen. Carbon. Oxygen. 
 
 42-85 : 57-14 :: i : 1-33 
 and 27-27 : 72-72 :: i : 2-66 
 
 Then plainly 
 
 1-33 : 2-66 :: i : 2 
 
 In other words, the proportion of oxygen by weight in 
 carbonic anhydride gas is double that contained in carbonic 
 
Daltorfs Atomic Theory. 85 
 
 oxide. Obviously the same fact may be expressed in 
 another way by taking the oxygen as unity instead of the 
 carbon : 
 
 Oxygen. Carbon. Oxygen. Carbon. 
 
 57-14 : 42-85 :: i : 750 
 7272 : 27-27 :: i : -375 
 
 Then 
 
 7S : '375 :: 2 : i 
 
 Or we may say that carbonic anhydride contains half as 
 much carbon as carbonic oxide. Similarly it will be found 
 that the ratios of the oxygen in the several oxides of nitro- 
 gen, of sulphur, and of many of the metals are as i, 2, 3, 
 c., to i, 2, 3, &c., of the other constituent. 
 
 Looking about for an explanation of these observations of 
 his, it occurred to Dalton to resuscitate the ancient theory 
 of the limited divisibility of matter. According to this 
 view, matter is made up of definite small parts or particles, 
 called ; atoms,' because it was assumed that they could not 
 be in any way cut or divided. This assumption of finite 
 particles as an explanation of the phenomena of chemical 
 combination thus became the basis of the modern theory of 
 molecules. But Dalton's term * atom ' was applied indiscrimi- 
 nately to the ultimate particles of compound as well as 
 elementary bodies. The relative weights also assigned to 
 these atoms were inferred from the results of the analysis 
 of only a few compounds. Consequently the word 'atom,' 
 in the Daltonian system, may be interpreted sometimes 
 'molecule,' sometimes 'atom,' and in other cases it cor- 
 responds to the term * equivalent,' as those words are now 
 understood by chemists. 
 
86 
 
 Chemical Philosophy. 
 
 CHAPTER XL 
 
 EQUATIONS. 
 CLASSIFICATION OF REACTIONS. 
 
 Equations. Chemical changes involve neither the de- 
 struction nor the creation of matter, but simply a redistri- 
 bution of the materials of which the acting masses are 
 composed. 
 
 In order, therefore, to represent symbolically the results 
 of any given action it is only necessary to write down the 
 formulae of the bodies engaged, and then to transpose their 
 symbols in such a manner as to build up the formulae of the 
 bodies which are produced. We thus arrive at equations in 
 which the signs + , , and = are employed, so far as the 
 weights of matter are concerned, in the same sense as in 
 algebra. 
 
 The following examples will serve to show the manner 
 in which chemical equations are to be read, as well as the 
 mode of investigating the relative weight of the bodies which 
 are formed or decomposed. The student should practise 
 reading equations aloud, or writing them out at full length, 
 according to these instructions : 
 
 Equation given, 
 
 2 HgO = 2Hg + O 2 . 
 
 This means that two molecules, or 432 parts, by weight, 
 of mercuric oxide yield two molecules, or 400 parts, by 
 weight, of metallic mercury, and one molecule, or 32 parts, 
 by weight, of oxygen. 
 
 Young students, however, would do well to ensure 
 precision by filling up such a scheme as the following : 
 
 
 2HgO 
 
 = 
 
 2Hg + 2 
 
 No. of molecules 
 
 2 
 
 give 
 
 2 and I 
 
 Name 
 
 Mercuric oxide 
 
 
 Mercury 
 
 Oxygen 
 
 Wt. of I mol. . 
 
 200 -f 16 = 2l6 
 
 
 200 
 
 32 
 
 Whole weight . 
 
 used 432 
 
 
 obtained 400 
 
 32 
 
Equations. 87 
 
 When gases or vaporisable bodies occur in such an 
 equation we can express the volume of each by recollecting 
 that, according to the law of Avogadro, the volumes of 
 gaseous molecules under like conditions are equal. We may 
 also connect together the weight and volume of a gas by 
 applying a rule which will be more fully explained in the 
 next section, namely, that the relative density of a body in 
 the gaseous state l is the half of its molecular weight. The 
 weight of a molecule of carbonic anhydride, for example, 
 is 44 : its relative density is, therefore, 22, and the weight 
 of a litre of the gas is 22 times as great as the weight of a 
 litre of hydrogen measured at the same temperature and 
 pressure. Now, the weight of a litre of hydrogen at o C., 
 and under a pressure equal to 760 mm. of the barometric 
 column, is '0896 gram or i crith. Hence we may say that 
 a litre of carbonic anhydride weighs 22 criths, or, if the 
 weight is to be expressed in grams, it will be 22 x -0896 
 grams. 
 
 Example. 
 
 C1 2 + H 2 = 2HC1- 
 
 A molecule of chlorine and a molecule of hydrogen 
 produce two molecules of hydrochloric acid. 
 
 
 C1 2 + H 2 
 
 _ 
 
 2HC1 
 
 No. of molecules 
 
 i 
 
 i 
 
 
 2 
 
 Name 
 
 Chlorine 
 
 Hydrogen 
 
 
 Hydrochloric acid or 
 
 
 
 
 
 hydrogen chloride 
 
 Wt. of I molecule 
 
 71 
 
 2 
 
 
 36-5 
 
 Whole weight . 
 
 used 71 
 
 2 
 
 
 obtained 73 
 
 Volume 
 
 i 
 
 i 
 
 
 obtained 2 
 
 
 or 2 
 
 or 2 
 
 
 or 4 
 
 If, now, we choose to attach a concrete value to the 
 symbols, we express it thus : 
 
 C1 2 + H 2 = 2HC1 
 Weight 71 criths 2 criths 73 criths; 
 
 1 The density of hydrogen being taken as I. 
 
88 Chemical Philosophy. 
 
 or 
 
 71 X -0896 grams + 2 x -0896 grams = 73 x '0896 grams 
 Vol. 2 litres + 2 litres = 4 litres. 
 
 The student having thoroughly mastered the foregoing 
 examples, and worked some of the exercises given at the 
 end of the section, is now in a position to solve such pro- 
 blems as the following, which may be taken as represen- 
 tative : 
 
 1. How many pounds of zinc are required to make 500 
 pounds of the crystallised sulphate ? 
 
 Formula of one molecule of crystallised zinc sulphate, 
 Zn SO 4 . 7H 2 O. 
 
 Molecular weight 287. 
 
 The symbol of zinc, Zn = 65, occurs only once in the 
 formula of the sulphate. Hence, 
 
 In 287 Ibs. of zinc sulphate there are 65 Ibs. of zinc. 
 
 i Ib. -J- lb. of zinc. 
 
 207 
 
 /. 500 Ibs. ,, would require -j- X 500 Ibs. 
 
 Ibs. oz. 
 
 Ans . 65x500 : =II33f of zinc. 
 287 
 
 2. How many litres of oxygen (at standard temperature 
 and pressure) are obtained by heating 10 grams of potassic 
 chlorate ? 
 
 or more simply 
 
 KC10 3 = KC1 + iiO a 
 126-6 74-6 48. 
 
 Taking the potassium chlorate in criths, we should obtain 
 3 litres of oxygen gas. 
 
Problems. 89 
 
 Then 
 
 122*6 criths 
 or 122-6 x -0896 grams give 3 litres of oxygen. 
 
 i gram would give $ -, litres. 
 
 122-6 x -0896 
 
 .'. 10 grams 3 - litre. 1 
 
 122-6 X 0-896 
 
 Ans. * =- 2 -7 \ i litres, or 2731 c.c. 
 
 122-6 x -0896 
 
 CLASSIFICATION OF REACTIONS. 
 
 Chemical action may take place in a great many different 
 ways, but every known chemical change to which bodies 
 are subject may be referred to one or other of the following 
 five typical modes of action : 
 
 I. Combination of entire molecules. 
 Examples : 
 
 Hg 
 
 + C1 2 = 
 
 HgCl 2 . 
 
 Mercury. 
 
 Chlorine. 
 
 Mercuric chloride. 
 
 CO 
 
 + Cl, = 
 
 COC1 2 
 
 Carbonic oxide. 
 
 Chlorine. 
 
 Carbonyl chloride. 
 
 NH 3 
 
 + HC1 = 
 
 NH 4 C1 
 
 Ammonia. 
 
 Hydrochloric acid. 
 
 Ammonium chloride. 
 
 CaO 
 
 + OH 2 = 
 
 Ca(HO) 2 
 
 Calcium oxide. 
 
 Water. 
 
 Calcium hydroxide. 
 
 Pb0 2 
 
 + S0 2 = 
 
 PbS0 4 
 
 Lead peroxide. 
 
 Sulphurous anhydride. 
 
 Lead sulphate. 
 
 1 Taking -0896 gram as the weight of I litre of hydrogen, I gram 
 is the weight of 1 1 -16 litres. This number is sometimes useful. Thus, 
 in the example worked out in the text, the calculation would be a trifle 
 shorter, thus : 
 
 KC10 3 give 3 
 
 1 22 '6 grams 16 x 3 grams 
 
 or 11*16 x 3 litres. 
 Then 10 grams of chlorate would give 
 33-48xio = 
 
 122-6 
 
9O Chemical Philosophy. 
 
 II. Splitting up of a compound molecule into its element s^ 
 or into simpler molecules. 
 
 Examples : 
 
 CaCO 3 = CaO + CO 2 
 
 Calcium Calcium Carbonic 
 
 carbonate. oxide. anhydride. 
 
 H 2 SO 4 = H 2 O + SO 3 
 
 Sulphuric acid. Water. Sulphuric anhydride. 
 
 MnCl 4 = MnCl 2 + C1 2 
 
 Manganic Manganous Chlorine, 
 
 tetrachloride. chloride. 
 
 Hg 2 I 2 Hg + HgI 2 
 
 Mercurous iodide. Mercury. Mercuric iodide. 
 
 C 7 H 6 5 = C 6 H 6 3 + C0 2 
 
 Gallic acid. Pyrogallol. Carbonic anhydride. 
 
 Many familiar decompositions, which at first sight appear 
 to belong to this class, are in reality double decompositions, 
 as an examination of one or two cases will show. The 
 decomposition of mercuric oxide by heat, for example, 
 seems to consist in a simple resolution of the body into its 
 elements, 
 
 HgO = Hg + O. 
 
 But when we write the equation molecularly, 
 
 HgO + HgO = Hg + Hg + 2 , 
 
 we may see that the decomposition of one molecule of the 
 oxide necessitates the splitting up of another. In order to 
 form one molecule of oxygen and we believe one molecule 
 to be the smallest quantity of the element capable of inde- 
 pendent existence we must take two atoms of it from two 
 separate molecules of the oxide. 
 
 The same remarks apply to other cases, such as the 
 decomposition of potassium chlorate by heat. 
 
 III. Rearrangement of the atoms constituting a molecule 
 so as to give rise to a new body. 
 
Classification of Reactions. 9 1 
 
 Two examples of this mode of transformation may be 
 given here. 
 
 (NH 4 )CNO converted by heat into CO(NH 2 ) 2 . 
 
 Ammonium cyanate. Urea. 
 
 C 6 H,CH 3 HN converted by heat into C 6 H 4 CH 3 H 2 N. 
 
 Methyl-aniline. Toluidine. 
 
 The further explanation of this kind of change is post- 
 poned to the chapter on ' Isomerism ' (Sec. IV.). 
 
 IV. Single Displacement. In this kind of change an 
 atom, or group of atoms, contained in a molecule is dis- 
 placed by another atom or group. 
 
 Examples : 
 
 Zn + 2HC1 = ZnCl 2 + H 2 
 
 Zinc. Hydrochloric acid. Zinc chloride. Hydrogen. 
 
 MgBr 2 + C1 2 = MgCl 2 + Br 2 
 
 Magnesium bromide. Chlorine. Magnesium chloride. Bromine. 
 
 2C 5 H n ONO + S0 2 = (C 5 H n O) 2 S0 2 + 2NO 
 
 Amyl nitrite. Sulphur dioxide. Amyl sulphate. Nitric oxide. 
 
 Many cases of precisely the same character are not easy 
 to find. The fact is, the great majority of reactions belong 
 to the next class of double decompositions. Even some 
 which seem to be single decompositions must in strictness 
 be so considered ; thus, the decomposition of hydrochloric 
 acid by sodium must be represented in this manner : 
 
 NaNa + HC1 + HC1 
 or Na 2 + 2HC1 = 2NaCl + H,, 
 not Na + HC1 = NaCl + H." 
 
 V. Double Decomposition^ or Metathesis. This is by far 
 the most general mode of reaction. Two or more molecules 
 coming together exchange some of their constituents so as 
 to give rise to the same number, or to a greater number of 
 molecules. We may, for the sake of completeness, classify 
 double decompositions under three divisions : 
 
92 Chemical Philosophy. 
 
 1. Those in which one of the reacting bodies is an 
 element ; e.g. 
 
 Na 2 + 2OH 2 = 2 NaOH + H 2 
 
 Sodium. Water. Sodium hydroxide. Hydrogen. 
 
 2. Those in which both are compounds ; e.g. 
 Na 2 CO 3 + CaCl 2 = 2NaCl + CaCO 3 
 
 Sodium carbonate. Calcium chloride. Sodium chloride. Calcium carbonate. 
 
 3. Those which result in the formation of ' substitution ' 
 compounds. This kind of reaction is not essentially different 
 in its nature from cases i and 2, but substitution products 
 among carbon compounds constitute a class of bodies so 
 remarkable in their characters as to deserve special notice. 
 Some examples of their formation are given in this place, 
 more particular mention being reserved for a later chapter. 
 
 Examples : 
 
 CH, 
 
 Methane. 
 
 + C1 2 = CH 3 C1 + HC1 
 
 Chlorine. Chloromethane. Hydrochloric acid. 
 
 C 6 H 6 
 
 Benzene. 
 
 + N0 2 HO = C 6 H 5 N0 2 + 
 
 Nitric acid. Nitrobenzene. 
 
 H 2 O 
 
 Water. 
 
 C 6 H 6 - 
 
 Benzene. 
 
 f HO.S0 2 .HO=C 6 H 5 .S0 2 .OH + 
 
 Sulphuric acid. Benzene-sulphonic acid. 
 
 H 2 O 
 
 C 6 H 6 
 
 Phenol. 
 
 + NO.OH =C 6 H 4 (N.OH)O -f 
 
 Nitrous acid. Isonitrosophenol. 
 
 H 2 
 
 CHAPTER XII. 
 
 CHEMICAL COMPOUNDS DISTINGUISHED FROM MIXTURES. 
 
 WHEN two bodies unite together to form a chemical com- 
 pound, they merge so completely one into the other as to 
 be no longer recognisable by any physical character. The 
 properties of chemical compounds are always quite distinct 
 from those of their constituents, and in general have not 
 even a remote resemblance to them. In the glistening red 
 
Mixtures and Chemical Compounds. 93 
 
 crystalline powder called by chemists mercuric oxide and 
 vulgarly ' red precipitate,' no trace can be detected by the 
 eye, or by any other sense, of the liquid, volatile, silvery 
 metal mercury, and the colourless, gaseous oxygen into 
 which it is resolved by the action of heat. In water, again, 
 whether examined in the condition of solid ice, liquid 
 water, or vaporous steam, we should look in vain for any re- 
 semblance to the gaseous elements, hydrogen and oxygen, 
 of which it consists. Neither can we detect in the pro- 
 perties of water any that can be regarded as intermediate 
 between those of the two constituents, or such as we should 
 expect to find exhibited by a mixture in which each element 
 retained its independence. 
 
 The law of definite proportions furnishes another criterion 
 by which the character of a body under examination may be 
 judged as to its title to rank as a definite chemical species. 
 For when it is found that the elements contained in a 
 substance are united together in the ratio of their atomic 
 weights, the probability that they are chemically combined 
 is very great ; whilst if they are not present in such pro- 
 portion, the inference that the substance is a mixture is 
 certain. 
 
 In order to decide whether a given substance is a true 
 chemical compound or a mere mechanical mixture, various 
 considerations are employed by chemists, the nature of 
 which depends very much upon the circumstances of each 
 particular case. If a solid body is the subject of investiga- 
 tion, it is examined under a microscope, in order to see 
 if its appearance is uniform throughout ; or, if crystallisable, 
 it is recrystallised, and the crystals compared with those of 
 the original substance. If soluble in any liquid, it may be 
 treated with a quantity of the solvent insufficient to take up 
 the whole. The part dissolved, after getting rid of the 
 solvent by evaporation or otherwise, ought to agree in every 
 respect with the undissolved portion, if the original body is 
 one compound and not a mixture. 
 
94 Chemical Philosophy. 
 
 In the case of those liquids which are volatile, and which 
 bear the application of heat without decomposition, the 
 boiling-point should remain constantly at the same tempera- 
 ture during the distillation of the whole, and portions taken 
 from the retort and from the distillate in the receiver ought, 
 in the case of definite compounds, to correspond in density 
 and all other physical and chemical characters. 
 
 When the body to be examined is a gas, the action of 
 solvents is tried; and if, after such treatment, the relative 
 proportions of the ingredients are undisturbed, the body 
 may be regarded as probably consisting of one compound. 
 This may be confirmed by observing whether these propor- 
 tions agree with the combining weights of the elements 
 present The phenomena of gaseous diffusion are not un- 
 frequently useful in helping to decide whether the elements 
 in a given gas are chemically combined or mechanically 
 mixed. Other means of a mechanical nature may be re- 
 sorted to in special cases, and occasionally considerable 
 ingenuity is called for in devising methods suited to the 
 occasion. 
 
 One or two examples will render these matters more 
 intelligible to the student. We will select cases in which 
 the law of definite proportions would afford no assistance in 
 the solution of the problem. 
 
 Fifty-six parts of iron would unite with thirty-two parts 
 by weight of sulphur ; but the two elements may be mixed 
 together in the state of fine powder, without exerting upon 
 each other any chemical action whatever, the compound, 
 ferrous sulphide, which would be formed by their union in 
 these proportions being produced only when they are 
 strongly heated together. The mixture of these two bodies, 
 though it might be indistinguishable from the compound by 
 appealing to the proportions of the two ingredients, would 
 yet be easily recognised by such properties as the follow- 
 ing: Under a microscope particles of iron and particles of 
 sulphur would be visible; a magnet would withdraw the iron 
 
Mixtures and Chemical Compounds. 95 
 
 from the powder, and leave the sulphur ; carbon bisulphide 
 would dissolve the sulphur, but would not affect the iron; a 
 separation could be effected by merely stirring up in water, 
 when the iron, by reason of its greater density, would 
 sink quickly to the bottom, leaving the sulphur suspended; 
 diluted sulphuric acid poured upon the mixture would 
 evolve hydrogen gas. The chemical compound has a uniform 
 appearance under the microscope ; if reduced to powder, it 
 could not be divided into two different portions by the use 
 of a magnet, by any solvent, or by elutriation with water ; 
 and, lastly, the action of diluted sulphuric acid would result 
 in the evolution of hydrogen sulphide gas, easily distinguish- 
 able from hydrogen by its odour, by its solubility in water, 
 and by many other properties. 
 
 The domain of organic chemistry supplies numerous 
 problems of the kind we are considering. 
 
 In certain kinds of tartar there exist the potassium salts 
 of two acids, which have the same composition, but some- 
 what different properties. These acids are called respec- 
 tively tartaric and racemic acids. The former rotates the 
 plane of polarisation of a ray of polarised light to the right, 
 whilst the latter is optically inactive. Crystals of tartaric 
 acid are permanent in the air, whilst those of racemic acid 
 contain a molecule of water, which, escaping at ordinary 
 temperatures, renders the crystals efflorescent. Racemic 
 acid and calcium racemate are decidedly, though not very 
 greatly, less soluble than the corresponding tartaric acid 
 and calcium tartrate. Hence it will be perceived that 
 whilst it is perfectly easy to distinguish the pure acids from 
 each other, the mere estimation of the amount of carbon 
 and hydrogen, or even an examination of a great many of 
 the salts, would not suffice to decide between the one or 
 the other of them and a mixture of the two. 
 
 A mixture of alcohol, water, and ether might be made 
 in such proportions (46 : 18 : 74) that it would possess 
 exactly the same composition as alcohol. But such a mix- 
 
96 Chemical Philosophy. 
 
 ture would be at once distinguished from alcohol by its 
 peculiar odour, and by separating into two layers on addi- 
 tion of water. When distilled, it would be found to boil 
 at a much lower temperature than the boiling-point of 
 alcohol ; and after about three-fourths had passed over into 
 the receiver, the liquid left behind in the retort would no 
 longer smell of ether, and would be easily recognised as 
 weak alcohol. 
 
 Again, the commercial liquid alkaloid toluidine is an 
 oil which, when distilled, boils steadily at about 200, and 
 no difference of composition can be detected between the 
 first portions of the distilled liquid and the last. And yet 
 this substance is a mixture of two alkaloids of the same 
 composition, one liquid, the other solid, the boiling-points 
 of which (200 and 198) differ so slightly that they cannot 
 be separated by any kind of fractional distillation. The 
 only process by which they can be completely separated de- 
 pends upon the fact that the oxalate of solid toluidine is 
 much less soluble in ether than the oxalate of the liquid 
 base. By patient repetition of this treatment with ether two 
 salts are obtained, which, when recrystallised any number 
 of times, undergo no further change in crystalline form and 
 solubility. In this condition they are believed to be pure 
 and homogeneous. 
 
 Examples of mixtures of gases, presenting the same 
 composition as true chemical compounds, might be easily 
 multipled. 
 
 Thus, equal volumes of hydrogen and chlorine constitute 
 a gaseous mixture which exhibits the colour and bleaching 
 action of chlorine ; and after shaking up with" solution 
 of soda, just one half its bulk of colourless inflammable 
 hydrogen remains. Hydrochloric acid gas, which contains 
 the same elements combined in the same proportions, is, 
 on the contrary, a colourless gas which no longer possesses 
 the bleaching power of chlorine, and is readily and com- 
 pletely soluble in water or in solution of soda. 
 
Atmospheric Air a Mixture. 97 
 
 Ethane and hydrogen gases in equal volumes furnish a 
 mixture which would be indistinguishable from marsh-gas 
 by ordinary quantitative analysis. But recollecting that 
 ethane is more than four times more soluble in water than 
 hydrogen, whilst the rate of diffusion of hydrogen is nearly 
 four times that of ethane gas, it would not be difficult to 
 distinguish the mixture from the compound. Further assist- 
 ance might be derived from a study of the action of chemical 
 agents upon the two gases, 
 
 The case of atmospheric air is one of so great import- 
 ance, that its consideration demahds some attention in this 
 place. 
 
 Neglecting accidental constituents, as well as the water 
 vapour and carbonic dioxide, which are always present, the 
 analysis of air from various localities has led to the conclu- 
 sion that it consists almost uniformly of 20-9 volumes of 
 oxygen with 79-1 volumes of nitrogen. The question 
 whether these two elements are united together chemically 
 has been decided in the negative, in accordance with such 
 considerations as the following : 
 
 1. The most accurate analyses seem to indicate that the 
 proportion of oxygen to nitrogen in the atmosphere is not 
 absolutely uniform, as would be the case if it were a 
 compound. 
 
 2. The quantities of oxygen and nitrogen present do not 
 bear any simple relation to the combining weights of those 
 elements. 
 
 3. When oxygen and nitrogen are mixed together they 
 show no signs of chemical action by evolution of heat or 
 contraction of volume ; and such a mixture, when due pro- 
 portions are employed, resembles atmospheric air in every 
 respect. 
 
 4. Water dissolves the constituents of the air in unequal 
 proportions, so that, by reason of the greater solubility of 
 oxygen, the air which may be expelled from common water 
 by boiling contains a larger proportion of that element 
 
98 Chemical Philosophy. 
 
 than is present in atmospheric air. 1 A chemical compound 
 would dissolve, as a whole, without change of composition. 
 
 5. When the rays emitted from a given source of heat 
 are transmitted through different gases, it is found that 
 compounds absorb a much larger amount than elementary 
 gases or mixtures of elementary gases. Thus the amount 
 of radiant heat absorbed by nitrous oxide, a colourless and 
 transparent gas, is more than 350 times as great as the 
 amount absorbed by a column of equal length of oxygen, 
 nitrogen, or atmospheric air. Between the absorbent 
 powers of the last three gases no difference can be detected, 
 and the natural inference, therefore, is that they are similarly 
 constituted. 
 
 6. The elements may be separated to a certain extent by 
 the mechanical process of diffusion through a porous plate, 
 called, in this case, atmolysis. 
 
 It has also been discovered by Graham (see Chap. V., 
 p. 44) that gases have the power of penetrating thin sheets 
 of india-rubber, and that the rate at which oxygen passes 
 through this material is more than two and a half times 
 that of nitrogen. Upon this observation he has based a 
 very instructive experiment, which proves conclusively the 
 fact that the oxygen of atmospheric air is not combined 
 with the nitrogen. An air-tight india-rubber bag is ex- 
 hausted as completely as possible by the Sprengel air-pump. 
 When the exhaustion is nearly perfect, it is found that gas 
 can still be slowly extracted from the bag by continuing the 
 operation ; and this gas is found by analysis to consist of 
 a mixture of nitrogen and oxygen, containing upwards of 
 forty per cent, of the latter ingredient. The gases thus 
 withdrawn from the bag result from the passage of the 
 gases of the atmosphere through the india-rubber partition ; 
 the oxygen, however, more rapidly than the nitrogen. The 
 explanation of this dialytic passage of gases through the 
 apparently impermeable caoutchouc appears to be that the 
 1 See also Chapter II. , p. 28, Solubility of Gases. 
 
Chemical Names. 99 
 
 gases are absorbed by the external film of that material, 
 that they penetrate in this condition to the other side of 
 the sheet, where evaporation occurs in consequence of 
 exposure to an atmosphere of very slight density. 
 
 CHAPTER XIII. 
 
 NOMENCLATURE. 
 
 A NAME may be used either for the purpose of indicating 
 some particular person or object, or it may serve to point 
 out relationships and to define the position which a thing 
 holds in some system of classification. 
 
 In the early days of chemistry the number of different 
 bodies known was comparatively small, and mere indicative 
 names fulfilled all the requirements of the time. But 
 when chemical research began to be regularly followed 
 and crowds of new compounds were constantly presenting 
 themselves, it became necessary to devise names which 
 would serve, not merely to distinguish one compound from 
 another, but to indicate, at least in some degree, the 
 relationship subsisting between allied bodies. The first 
 attempts of this kind were naturally imperfect, the devices 
 employed being wholly inadequate to the requirements of 
 the case. For example, it soon became evident that the 
 mere employment of adjectives, as in the names blue, green, 
 and white vitriol, and the like, could have only a very 
 limited application. 
 
 It was only after oxygen had been discovered, and its 
 compounds were called oxides by Lavoisier, that chemical 
 names began to assume some appearance of precision. The 
 nomenclature adopted by the leading chemists of that 
 period has met with very general acceptance ; and, although 
 modified in detail, the system of the present day is based 
 essentially upon the same principle. The names now 
 
 H 2 
 
TOO Chemical Philosophy. 
 
 employed by chemists for scientific purposes sometimes 
 assume rather formidable dimensions, but, unlike the fanciful 
 names in use in connection with some branches of natural 
 history, every syllable has a significance of its own. And, 
 in spite of their length and frequent uncouthness, it may 
 fairly be claimed for these names that they are in few cases 
 inconvenient practically, whilst they do very fairly realise 
 the idea originated by Lavoisier, namely, that of represent- 
 ing, as by a formula, the composition of the bodies for which 
 they stand. 
 
 Nevertheless, many of the old names dating their origin 
 from the times of the alchemists have become, by long 
 familiarity, so incorporated into the language of medicine, 
 of commerce, and the arts, that it is neither possible nor 
 advisable for the chemist to reject their use. In the majority 
 of cases, indeed, they are very serviceable, and no chemical 
 student should disdain to make himself acquainted with 
 s-uch terms as caustic potash, alum, borax, or oil of vitriol 
 wood-spirit or marsh-gas, and to use them when they ex- 
 press all that is required, alternatively with the more formal 
 and systematic names, the employment of which, for certain 
 purposes, has been rendered necessary by the advance of 
 knowledge. 
 
 Names of Elements. Several of the metals iron, copper, 
 lead were known in very early times. Those elements 
 which have been discovered by more modern chemical re- 
 search have received names which in some cases recall their 
 origin, e.g. silicon (silex, flint) ; whilst others were sug- 
 gested by some prominent characteristic, e.g. chlorine 
 (xAxopos, green), or by their chemical relations, real or sup- 
 posed, e.g. oxygen (o|vs, acid ; yewaw, I generate). 
 
 Metals which have been discovered in modern times 
 have all been designated by Latinised names, with the ter- 
 mination ' him ' * potassium, sodium, lithium, thallium, 
 
 1 Unfortunately this termination has been erroneously applied to the 
 name of the non-metal selenium, from a belief entertained at the time of 
 
Names of Binary Compounds. 101 
 
 &c. ; whilst the names of the non-metallic elements are 
 characterised by no syllable which is common to them all. 
 
 Names of Binary Compounds. When two elements are 
 combined together, the last syllable or two of the name of 
 one of them is changed into the suffix ide. Thus, we 
 have CaO, calcium oxide ; HC1, hydrogen chloride ; PbS, 
 lead sulphide. In nearly all cases it is the name of the 
 more negative or chlorous element which is thus modified, 
 the name of the metal or corresponding element remaining 
 unaltered. 
 
 Prefixes from the Greek mono-, di-, tri-, &c. serve to 
 indicate the number of atoms of the chlorous element 
 present in a molecule of the compound. Thus 
 
 N 2 O is nitrogen monoxide. 
 
 N 2 O 2 nitrogen dioxide. 
 
 N 2 O 3 nitrogen trioxide. 
 
 N 2 O 4 nitrogen tetroxide. 
 
 N 2 O 5 nitrogen pentoxide. 
 
 Not unfrequently, however, when the actual number of 
 atoms is doubtful, or when it is desired simply to indicate a 
 relation between two compounds, that which contains the 
 larger proportion of the chlorine or oxygen or similar 
 substance is distinguished by the suffix if, whilst the other 
 ends in ous. Thus, in the foregoing series, we may dis- 
 tinguish- Na0 a nitRwj oxjde> 
 
 N 2 O 2 , or more correctly NO, as nitr/V oxide. 
 
 Similarly 
 
 N 2 O 3 is mtrous anhydride. 
 N 2 O 5 is nitnV anhydride. 
 
 A few other marks serve in special cases ; thus, Fe 2 O 3 , 
 
 its discovery that it was a metal. I have ventured in these pages to 
 change the final syllable into on, selenion. The alteration of the name 
 silicium into silicon will, I hope, be considered sufficient precedent. 
 
IO2 Chemical Philosophy. 
 
 Cr 2 O 3 , Mn 2 O 3 , A1 2 O 3 , are often called ji^a/oxides. 1 When 
 there is a series of oxides, chlorides, or sulphides of the 
 same elements, the syllables froto 2 and/^r 3 are sometimes 
 prefixed to the name to indicate respectively the poorest 
 and the richest in oxygen, chlorine, or sulphur, as the case 
 may be. Thus, the two oxides of iron, which are usually 
 represented by the formulae FeO and Fe 2 O 3 , may be dis- 
 tinguished by one or other of the following pairs of names, 
 according as it is desired simply to indicate that the ratio of 
 oxygen to iron is greater in the one compound than in the 
 other ; or to imply, what is a matter of far less certainty, 
 that so many atoms of the two elements form one molecule 
 of the compound. 
 
 FeO, Ferrous oxide : Iron protoxide ; Iron monoxide ; 
 
 or or 
 
 Protoxide of iron. Monoxide of iron. 
 
 Fe 2 O 3 , Ferric oxide : Iron peroxide ; Iron sesquioxide ; 
 
 or or 
 
 Peroxide of iron. Sesquioxide of iron- 
 
 Names of Acids and Salts. The same principles guide 
 the construction of the names of these compounds, and a 
 single example will go far towards explaining their applica- 
 tion. It happens, in the case of chlorine, that an unbroken 
 succession of compounds is formed by the union of this 
 element with hydrogen and oxygen. These are their names 
 and formulae : 
 
 HC1 . . . Hydrochloric acid. 
 
 HC1O . . . Hypochlorous acid. 
 
 HC1O 2 . . . Chlorous acid. 
 
 HC1O 3 . . . Chloric acid. 
 
 HC1O 4 . . . Perchloric acid. 
 
 Here, again, the terminations ous and ic serve to indicate 
 
 1 Latin, sesqtti, one and a half. 2 Greek, irpuros, the first. 
 
 3 Greek, vvcp, above, over, exceeding. 
 
Names of Acids and Salts. 103 
 
 different grades of oxidation, whilst the prefixes hypo and 
 per respectively announce a smaller and a larger amount of 
 oxygen than is contained in the chlorous and chloric acids. 
 In the name hydrochlor/V acid, it is evident that the termi- 
 nation has nothing to do with the presence or absence of 
 oxygen, and this use of the termination forms an exception 
 to the general rule. If we consent to regard acids as salts 
 of hydrogen, we may write names for this series of com- 
 pounds which are constructed in all respects in the same 
 manner as those which are applied to salts in general. 
 
 Instead of We may write 
 
 Hydrochloric acid . . Hydrogen chloride, 
 
 Hypochlorous acid . . Hydrogen hypochlorite. 
 
 Chlorous acid . . . Hydrogen chlorite. 
 
 Chloric acid . . . Hydrogen chlorate. 
 
 Perchloric acid . . . Hydrogen perchlorate. 
 
 And here another rule must be attended to, namely, 
 that when the name of a salt ends in tie, the name of the 
 acid or hydrogen salt with which it corresponds terminates 
 in ous, whilst salts in ate are derived from acids whose 
 names end in if. 
 
 One more example will serve to emphasise this rule : 
 
 (H 2 SO 3 Hydrogen sulphite, or sulphurous acid. 
 
 NaHSO 3 Sodium-hydrogen sulphite, 
 
 or acid sulphite of sodium. 
 Na 2 SO 3 Sodium sulphite, 
 
 \ or disodium sulphite (neutral). 
 
 H 2 SO 4 Hydrogen sulphate, or sulphuric acid. 
 
 NaHSO 4 Sodium hydrogen sulphate, 
 
 or acid sulphate of sodium. 
 Na 2 SO 4 Sodium sulphate, 
 
 or disodium sulphate (neutral). 
 
IO4 Chemical Philosophy. 
 
 NAMES OF CARBON COMPOUNDS. 
 
 Dr. Hofmann, a few years ago, made a proposition with 
 the object of reducing to some degree of order the con- 
 fused nomenclature of the very numerous compounds of 
 carbon and hydrogen. The number of atoms of carbon is 
 indicated by incorporating the Greek numeral into the name 
 and introducing a vowel into the last syllable, in order to 
 show the proportion of hydrogen. In the following table 
 the words * methane, ethane, propane,' are based upon the 
 names of the radicles * methyl, ethyl, propyl,' which have 
 long been in use and are too familiar to be discarded : 
 
 CH 4 CH 2 
 
 Methane. Methene. 
 
 CTT (~< TT r* TT 
 
 2 JTlg V^ 2 -d. 4 ^oAJ-2 
 
 Ethane. Ethene. Ethine. 
 
 C 3 H 8 C 3 H 6 C 3 H 4 C 3 H 2 
 
 Propane. Propene. Propine. Propone. 
 
 CTT r^TJ r~* IT f 1 T-T C* 13 
 
 4 tl 10 ^4^8 ^4^6 '^4 Jtl 4 ^4 Jtl 2 
 
 Tetrane. Tetrene. Tetrine. Tetrone. Tetrune. 
 
 CTT f~* TT fl TT (~* TT f~* TT f* TT 
 
 S**-ll ^5^-10 *~'*"8 ^*5**6 ^5- n 4 ^5 n 2 
 
 Pentane. Pentene. Pentine. Pentone. Pentune. 
 
 The succeeding terms would run : hexane, heptane, 
 octane, enneane, decane, &c. 
 
 The radicles formed from these bodies by loss of 
 hydrogen are furnished with the termination yl\ thus CH 3 , 
 derived from methane by removing H, is called methyl ; CH 
 is methenyl ; C 2 H 5 ethyl, and so on. 
 
 Unfortunately, the names which have long been applied 
 to a great many carbon compounds are in the majority of 
 cases fanciful, and, even in allied compounds, have no rela- 
 tion or resemblance to one another. They generally bear 
 some reference either to the original source of the body, or 
 to some more or less prominent characteristic. Thus, formic 
 acid is so called because it was originally obtained from the 
 bodies of ants (formica, an ant), acetic acid because it was 
 
Names of Carbon Compounds. 105 
 
 procured from vinegar (acetum). In like manner succinic 
 acid is the acid obtained by distillation of amber (sucrinum\ 
 and lactic acid from sour milk (lac}. 
 
 Attempts have been made to introduce some degree of 
 system into this crowd of heterogeneous materials by restrict- 
 ing the use of certain terminations. Thus the hydrocarbons 
 of the aromatic series have names which all end in em ; thus, 
 benzene, toluene, anthracene. Alcohols and bodies re- 
 sembling them claim the terminal syllable ol ; e.g. carbinol, 
 phenol, thymol. Basic nitrogenous bodies are represented by 
 names ending in ine ; thus, ethylamine, quinine, strychnine. 
 
 These conventions do serve their purpose to some extent, 
 though the whole matter is in a condition far from satis- 
 factory. 
 
 EXERCISES ON SECTION II. 
 
 1. Give the names of the elements represented by the following 
 symbols : Al, Sb, Fe, Mg, Hg, Mn, Ca, C, Cl, I, N, P, K, S, Ag, 
 Na, Br, Cu, F, H, Pb, O, Si, Zn. 
 
 2. Write down the symbols and atomic weights of 
 
 Barium, boron, bromine ; 
 
 Calcium, carbon, copper ; 
 
 Magnesium, manganese, mercury, silver ; 
 
 Phosphorus, potassium, lead ; 
 
 Sulphur, sodium, silicon ; 
 
 Iron, iodine, chlorine, oxygen, nitrogen. 
 
 3. Read these symbols and formuke, thus : 
 
 N., represents one molecule of nitrogen, consisting of two atoms ; 
 O, O 2 , OH 2 , 2OH 2 , HC1, H 2 , C1 2 , NH 3 , H 3 PO 4 , H 2 SO 4 , FeSO 4 , 
 2FeS0 4 , A1,(S0 4 ) 3 , I2~OH 2 , I2A1 2 (SO 4 ) 3 , CO 2 , 3CO 2 . 
 
 4. Write down the formulae and molecular weights of water, am- 
 monia, hydrochloric acid, carbonic anhydride, sulphuric acid, ferrous 
 sulphate, aluminic sulphate, phosphoric acid. 
 
 5. Write down the whole weight represented by each of the 
 following expressions : 2HgO, ioOH 2 , sFeS, 3FeS 2 , 2CS 2 , 
 KC 4 H 5 O ti , K 2 C 4 H 4 O 6 , 5C 7 H 9 N, I2CH 4 , KA1(SO 4 ) 2 + I2OH 2 , 
 3[NH 4 Cr(S0 4 ) 2 +i20HJ. 
 
Io6 Chemical Philosophy. 
 
 6. Name the following compounds : BaO, CaO, MgO, ZnS, 
 KC1, NaBr, AgF, H 2 S, HI, KCN, SSe, BN, H 3 P. 
 
 7. BaO, BaO 2 ; Hg.O, HgO ; FeS, FeS.> ; MnO, Mn,O 3 , 
 Mn0 2 ; FeO, Fe,O 3 > Fe 3 O 4 ; N 2 O, N 2 O 2 , N,O W N,O 4 , N,O 5 ; 
 P.,S 3 , P 2 S 5 ; SnCL, SnQ 4 ; FeBr 2 , Fe 2 Br 6 ; Cu.Cl 2 , CuCl 2 ; 
 CrCl 2 , Cr 2 Cl 6 , CrF 6 ; SbBr 3 , SbBr 5 . 
 
 8. KNO,, KNO 3 (ate) ; K 2 SO 3 , K.,SO 4 (ate) ; KC1, KC1O, 
 KC10 2 , KC10 3 (-ate), KC1O 4 ; KI, KIO 3 (-ate), KIO 4 ; 
 NaHS0 3 , Na,S0 3 ; Na,HPO 4 , Na 3 PO 4 , NaH 2 PO 4 ; H 3 PO 2 , 
 H,PO W H 3 P0 4 (-ic); HC1, HC1O, HBrO, HC1O,, IIC1O 3 , HIO 3 , 
 HC1O 4 , HBrO 4 . 
 
 9. Write out the following equations according to the scheme on 
 p. 86 : 
 
 (a} 
 
 MnO 2 
 
 + 4HC1 
 
 = 
 
 MnCl 2 - 
 
 f- C1 2 + 
 
 2H 2 0. 
 
 (b} 
 
 2KI 
 
 + C1 2 
 
 
 
 2KC1 
 
 f I, 
 
 
 (0 
 
 S0 2 + 
 
 2OH 2 + C1 2 
 
 = 
 
 H 2 S0 4 
 
 f 2HC1. 
 
 
 (^) 
 
 NaN0 3 
 
 + H 2 S0 4 
 
 = 
 
 NaHSO 4 
 
 f HNO 3 . 
 
 
 (e} 
 
 2MnO 2 
 
 + 2H 2 S0 4 
 
 = 
 
 2MnSO 4 - 
 
 h 2H 2 O + 
 
 2 . 
 
 (f. 
 
 ) 2K2Cr 2 O 7 
 
 + 8H 2 S0 4 
 
 = 
 
 2K 2 S0 4 H 
 
 h 2Cr 2 (S0 4 ) 3 
 
 + 8H 2 
 
 10. Write out the following equations according to the scheme on 
 p. 87:- 
 
 () 2OH 2 + 2C1 2 = 4HC1 + O 2 . 
 
 (b) CO 2 + C (solid) = 2CO. 
 
 (c) 2CO + O 2 2CO 2 . 
 (^/)2NH 3 = N 2 + 3H 2 . 
 (e) 2NH 3 + 3C1 2 = N 2 + 6HC1. 
 (/) ^H 4 N0 3 (solid) = NX) + 2H 2 0. 
 
 11. Write out in symbolic equations 
 
 (a) Ammonium chloride and calcium hydroxide give ammonia, cal- 
 cium chloride, and water. 
 
 (b} Ammonium nitrite (heated) yields nitrogen and water. 
 
 (c ) Common salt and sulphuric acid yield sodium, hydrogen sulphate, 
 and hydrochloric acid. 
 
 (d) Copper and nitric acid yield copper nitrate, nitric oxide, and 
 water. 
 
 (e) Mercury and sulphuric acid yield mercuric sulphate, sulphurous 
 anhydride, and water. 
 
 (/) Antimonious sulphide and hydrochloric acid yield antimonious 
 chloride and sulphuretted hydrogen. 
 
 12. How many grams of oxygen are required to burn 24 grams of 
 carbon and 32 grams of sulphur ? 
 
Exercises on Section II. 107 
 
 13. How many pounds of zinc are there in 350 pounds of the sul- 
 phate ZnSO 4 ? 
 
 14. How much sulphur will give 100 kilograms of sulphuric acid, 
 H 8 S0 4 ? 
 
 / 15. How many pounds of black oxide of manganese ate required to 
 yield, by the action of hydrochloric acid, 112 pounds of chlorine? 
 
 / 16. How many pounds of chalk containing 96 per cent, of calcium 
 carbonate, CaCO 3 , will neutralise 250 pounds of sulphuric acid ? 
 
 (In the following examples the gases are supposed to be at normal 
 temperature and pressure.) 
 
 17. Find the weight of 20 litres of oxygen, of 50 litres of chlorine, 
 of 250 litres of ammonia. 
 
 1 8. How many litres of oxygen are required to combine with 
 o. 12 criths of carbon ? fi. 2 grams of sulphur ? 
 
 7. 10 grams of carbon ? 
 
 " 19. How many litres of chlorine are required to decompose 12 litres 
 of hydriodic acid ? 
 
 2HI + C1 2 = 2HC1 + I 2 
 
 20. How many litres of chlorine are required for the complete 
 decomposition of 10 litres of olefiant gas ? 
 
 C 2 H 4 + 2C1 2 = C 2 + 4HC1 
 
 21. How many litres of hydrogen are obtained by dissolving 12 
 grams of magnesium in an acid ? 
 
 22. What weight of potassium chlorate is required to yield 35,000 
 cubic centimetres of oxygen ? 
 
 23. What materials and what quantities would you employ in order 
 to obtain 50 litres of each of the oxides of carbon ? 
 
 24. How much mercuric cyanide, Hg(CN) 2 , must be used to furnish 
 50 c.c. of cyanogen, C 2 N 2 , assuming that 60 per cent, of the cyanogen 
 is obtained in the gaseous form ? 
 
 25. Red oxide of copper contains 88-8 parts of copper and 11-2 
 parts of oxygen by weight ; black oxide of copper contains 79*87 of 
 copper and 20-13 of oxygen. If the formula of the black oxide is 
 CuO, how should the red oxide be represented ? 
 
 26. Water contains 88-8 of oxygen, ii-i of hydrogen. If its 
 formula is OH 2 , find the formula for peroxide of hydrogen, which 
 contains 94-12 O and 5-88 H. 
 
 27. Two hydrocarbons have the following composition : 
 
 I. II. 
 
 Carbon . . 8571 ) _ CR 92-3 
 
 Hydrogen . . 14 '29/ ~ 77 
 Find the formula for II. 
 
loS 
 
 Chemical Philosophy. 
 
 28. Show that the composition of the oxides of nitrogen, of man- 
 ganese, and of chromium are in accordance with the law of multiple 
 proportion. 
 
 29. Examine the following equations, attach the name to each 
 formula, and classify the reactions according to Chapter XL , pp. 89 to 
 92, giving reasons in doubtful cases : 
 
 (a) Na 2 + 20H 2 
 
 (b) Fe + H 2 SO 4 .H.X) 
 
 (c) (electrolysis) 2HC1 
 
 (d) so 2 
 
 + HA 
 
 () (heat) 
 
 + 3MnO 2 
 
 (/) OKH 
 
 + HC1 
 
 fr) 2 H 2 
 
 -i- O 3 (ozone) 
 
 (A) S0 3 
 
 + H 2 O 
 
 (0 PH 3 
 
 + HI 
 
 ( NA 
 
 + H 2 
 
 (k) PA 
 
 + 3H0 
 
 (/) 2KI 
 
 -i- C1 2 
 
 () QH 4 
 
 4- Br 2 
 
 Ethj'lene. 
 
 
 () NaNO 3 
 
 4- H 2 S0 4 
 
 () (heat) 
 
 2 KC10 3 
 
 (/) Mn0 2 
 
 4- S0 2 
 
 (y) Mn0 2 
 
 + 2SO 2 
 
 (r) C 2 H 4 2 
 
 + 3C1 2 
 
 (j) KjS 
 
 + CS 2 
 
 (0 C fi H 5 HO 
 
 + HNO, 
 
 Phenol. 
 
 
 () C 2 H 4 2 
 
 + PC1 5 
 
 (*) C 10 H 18 
 
 + HC1 
 
 (w/) 3HC1 
 
 + HN0 3 
 
 (*) As 2 3 
 
 + 2HNO 3 + 2l 
 
 O) (heat) 
 
 NaNH 4 HP0 4 
 
 (2) (heat) 
 
 2Na 2 HPO 4 
 
 20NaH 4 
 
 FeSO 4 .H 2 O 
 
 H 2 + C1 2 
 
 H 2 S0 4 
 
 Mn 3 O 4 + 
 
 KC1 + 
 
 OH 2 + 
 
 H 2 S0 4 
 
 PH 4 I 
 
 2HN0 3 
 
 2H 3 P0 4 
 
 2KC1 
 
 C 2 H 4 Br 2 
 
 H 2 
 -H 3 
 
 OH 2 
 
 20, 
 
 + I* 
 
 HNO 3 + NaHSO 4 
 = KC1 + KC10 4 
 = MnS0 4 
 
 MnS 2 O 6 
 = C 2 HC1A 4- 3HC1 
 
 K 2 CS 3 
 = C tt H 4 NO 2 HO 4- H 2 O 
 
 2 
 
 = C 2 H 3 OC1 + 
 
 POC1 3 
 
 4- HC 
 
 Acetyl chloride. 
 
 
 
 C 10 H 17 C1 
 
 
 
 = 2H 2 O 4- 
 
 NOC1 
 
 4- C1 2 
 
 = 2H,AsO 4 + 
 
 (NO + 
 
 N0 2 ) 
 
 = NaPO, + 
 
 NH 3 
 
 4- H,0 
 
 = Na 4 P 2 7 + 
 
 H.p 
 
 
 30. How many cubic centimetres of ammonia (measured at 15 and 
 under 740 mm.) would be obtained from 53! grams of ammonium 
 chloride ? 
 
 31. How many cubic centimetres of sulphur dioxide (measured at 
 20 and 740 mm. ) can be obtained by the action of copper on 20 grams 
 of sulphuric acid ? 
 
109 
 
 SECTION III. 
 
 CHAPTER XIV. 
 
 ATOMIC WEIGHTS. 
 
 THE first step towards the determination of the ratios, com- 
 monly called atomic weights, is to ascertain the proportions 
 in which the several elements unite with some one of their 
 number, the combining proportion of which is either taken 
 as the unit, or whose ratio of combination has already been 
 determined. With this object the form of experiment must 
 be suitably chosen, and every precaution taken against error, 
 which, however, can never be wholly avoided, even by the 
 most skilful and experienced operators. 
 
 Oxygen was formerly employed as the standard of com- 
 parison, and its atomic weight taken as 100. But, since 
 hydrogen is known to enter into combination in smaller 
 proportion than any other element, it is more convenient to 
 adopt it as the standard, and take its atomic weight as unity. 
 The most important relation to establish at the outset is the 
 proportion in which oxygen and hydrogen unite together to 
 form their most stable and characteristic compound, water. 
 This has been attempted by a number of experimenters, but 
 notwithstanding all the care and labour which has been 
 spent upon the investigation, it can hardly be said that the 
 values obtained represent so accurately the combining ratios 
 of these two elements that further experiment is unneces- 
 sary. Two chief methods have been adopted. Dumas, in 
 1842, published the result of his experiments, the first which 
 
no Chemical Philosophy. 
 
 were conducted with any near approach to accuracy. His 
 plan consisted in passing pure hydrogen through a weighed 
 bulb containing pure oxide of copper heated to redness, 
 collecting the water thus formed and weighing it. Assuming 
 the value i for the atomic weight of hydrogen, Dumas 
 obtained for oxygen the number 15 '9607. Erdmann and 
 Marchand by the same process obtained the value 15 '9733. 
 The second important method consists in determining 
 directly the relative densities of the two gases by weighing 
 them in large globes. By this mode of operating Regnault 
 found oxygen to be 15-9628 times heavier than hydrogen. 
 On reviewing the work that has been done by other experi- 
 menters, these values are substantially confirmed, and it 
 appears to be certain that the combining ratio of oxygen 
 is somewhat less than 16, the number commonly used, and 
 is probably very near to 15*9633. 
 
 The most trustworthy estimations of the combining 
 weight of carbon were made by the simple method of burn- 
 ing a weighed quantity of pure carbon, in the form of 
 graphite or diamond, in oxygen, and collecting the carbon 
 dioxide formed by means of caustic potash contained in 
 suitable absorption apparatus, carefully weighed. In carbon 
 dioxide 16 parts of oxygen combine with 6 parts of carbon 
 almost exactly; but assuming the value 15*9633 already 
 given for oxygen, the combining weight of carbon becomes 
 11-9736. 
 
 For the estimation of the combining proportions ot 
 other elements different methods have to be adopted. It 
 will be sufficient if two or three of these are indicated. 
 Chlorine, potassium, and silver may be taken as being both 
 important and representative. 
 
 By decomposing potassium chlorate by heat or by hydro- 
 chloric acid, and weighing the resulting potassium chloride, 
 experiments conducted at different times by Berzelius, 
 Penny, Marignac, and Stas led to results the general mean 
 of which gave 60-846 parts of potassium chloride from 100 
 
Atomic Weights. 1 1 1 
 
 parts of potassium chlorate. This may be expressed as 
 
 follows 
 
 follows : 
 
 Potassium chloride : oxygen :: 60-846 : 39*154, 
 or KC1 : O 3 :: 74-4217 : 47-8899. 
 
 This number, 74-4217, is the combining weight of po- 
 tassium chloride, and is made up of the combining weights of 
 potassium and chlorine. These must be separately estimated 
 as follows. ^Starting with pure silver, it may be converted 
 into chloride either by dissolving it in nitric acid and adding 
 hydrochloric acid, or by heating in chlorine gas. Experi- 
 ments of this kind by a number of chemists, among whom 
 Stas is conspicuous, have shown that 100 parts of silver 
 combine with 32*8418 parts of chlorine. It has also been 
 shown that 100 parts of silver dissolved in nitric acid require 
 for exact precipitation 69-1032 parts of chloride of potas- 
 sium. Now, these 69-1032 parts of potassium chloride 
 must contain 32*8418 parts of chlorine (which is the amount 
 required to combine with 100 parts of silver) and 36-2614 
 parts of potassium. Then, since 69*1032 parts of potassium 
 chloride correspond to 100 parts of silver, 
 
 KC1 : Ag :: 74*4217 : #, 
 
 and x is the combining ratio of silver = 107-696. 
 
 When the value already adopted for oxygen, namely, 
 15-9633, is taken, the combining value for silver deduced 
 from all the best experiments is 107*675. By similar calcu- 
 lations the number for potassium is 39*019, and for chlorine 
 35'37o. 
 
 Such numbers as these express, with the greatest approach 
 to accuracy at present possible, the proportion in which these 
 elements enter into certain of their most stable and definite 
 compounds, but the question whether they represent the 
 atomic weight, that is, the smallest proportions in which the 
 several elements enter into combination, can only be 
 answered by appeal to other considerations. The determi- 
 
112 Chemical Philosophy. 
 
 nation of the atomic weight, then, is accomplished by first 
 estimating as exactly as possible the combining proportion 
 of the element concerned, and then finding, by the applica- 
 tion of the methods now to be described, a factor by which 
 this value is to be multiplied or divided. In those cases in 
 which it is possible to apply several of the following rules, it 
 is found that the number indicated by the application of one 
 of them is identical with the number indicated by the others. 
 This is important as showing that when the atomic weights 
 of two elements have been determined by two different 
 methods, the probability that they are comparable with each 
 other and with the same standard amounts to practical cer- 
 tainty. Thus, the atomic weight of silver determined by 
 the specific heat method is equally probable with that of 
 carbon derived from the composition and density of its 
 vaporisable compounds, although silver forms no volatile 
 compound, and carbon exhibits anomalies in its specific 
 heat. 
 
 The atomic weight, as deduced from other considera- 
 tions, is represented in some cases by the same number as 
 the density of the element in the gaseous r state, the density 
 of hydrogen being taken as the unit. This is true of oxy- 
 gen, nitrogen, chlorine, bromine, iodine, sulphur, selenion, 
 and probably potassium and sodium. But the densities of 
 the vapours of mercury, cadmium, and zinc, as compared 
 with hydrogen, are respectively half the atomic weights, 
 whilst the vapour-densities of arsenic and phosphorus are 
 represented by numbers which are twice the atomic weights. 
 Hence the vapour-density of an element is no guide to the 
 determination of the atomic weight. 
 
 I. According to the system here adopted, and which will 
 be further dwelt upon in the next chapter, the bulk of one 
 part by weight of hydrogen is regarded as the volume of 
 the atom of that element, and is selected as the unit for 
 comparison of other volumes, atomic and molecular. Twice 
 this bulk of hydrogen contains a molecule, and all mole- 
 
Determination of Atomic Weights. 113 
 
 cules in the gaseous state occupy the same volume. Now, 
 according to the atomic theory, a molecule cannot contain 
 less than one atom of any element ; and, consequently, if 
 we ascertain what is the smallest quantity of an element 
 contained in a molecule of any compound of which it may 
 be a constituent, we shall have determined the atomic 
 weight of the element. 
 
 The atomic weight, then, may be said to be the smallest 
 weight of the element ever found in two volumes of the 
 vapour of any of its volatile compounds, the bulk of one 
 part by weight of hydrogen, at the same temperature and 
 pressure, being considered as one volume. 
 
 Suppose, for example, it is required to find by this rule 
 the atomic weight of oxygen, we have only to ascertain the 
 vapour densities of a number of compounds containing that 
 element, and the weight of oxygen contained in each. The 
 results are then tabulated in the following manner : 
 
 Volatile Compounds containing 
 Oxygen 
 
 Relative Density, 
 that is Weight, 
 of i Volume of 
 Gas or Vapour at 
 same Temp, 
 and Pressure 
 
 Weight of 
 Two 
 Volumes 
 
 Weight of 
 Oxygen con- 
 tained in 
 Two 
 Volumes 
 
 Water . 
 
 9 
 
 18 
 
 16 
 
 Carbonic oxide 
 
 H 
 
 28 
 
 16 
 
 Carbonic anhydride 
 
 22 
 
 44 
 
 32 
 
 Sulphurous anhydride 
 Sulphuric anhydride 
 
 32 
 40 
 
 64 
 
 80 
 
 32 
 48 
 
 Nitrous oxide 
 
 22 
 
 44 
 
 16 
 
 Nitric oxide 
 
 15 
 
 3 
 
 16 
 
 Alcohol . 
 
 23 
 
 5$ 
 
 16 
 
 Ether . 
 
 37 
 
 74 . 
 
 16 
 
 Acetic acid 
 
 3 
 
 60 
 
 3 2 
 
 &c. 
 
 &c. 
 
 
 &c. 
 
 Two volumes of the vapour of any volatile compound, 
 therefore, never contain less than 16 parts of oxygen, and 
 hence 16, or more accurately 15*96, is accepted as its 
 atomic weight. 
 
 This is a rule of very general applicability, for although 
 
Chemical Philosophy. 
 
 a great many of the elements, carbon for example, are quite 
 incapable of being volatilised at any manageable tempera- 
 ture, they yield a large number of easily volatile compounds. 
 
 There are, however, many metals which are neither 
 vaporisable by themselves nor when in union with other 
 elements. In such cases this rule cannot be applied, and 
 information has to be sought in a different direction. 
 
 II. Law of Dulong and Petit. ' The specific heats of 
 the solid elements are inversely proportional to their atomic 
 weights.' Whence it follows that the product of the multi- 
 plication of the specific heat by the atomic weight is a 
 constant number. 
 
 In the following table are given the specific heats of the 
 most important of the elements, together with their atomic 
 weights : 
 
 Name of the Element 
 
 Atomic 
 Weight 
 
 Sp. Ht. of 
 Equal Weights 
 
 Sp. Ht. of 
 Atomic Weights 
 
 Aluminium 
 
 27 
 
 2143 
 
 578 
 
 Antimony 
 
 120 
 
 0508 
 
 6-09 
 
 Arsenic . 
 
 75 
 
 0814 
 
 6-ii 
 
 Bismuth 
 
 208 
 
 0308 
 
 6 '40 
 
 Boron (crystallised) 
 
 ii 
 
 2300 
 
 2'53 
 
 Bromine (solid) 
 
 80 
 
 0843 
 
 674 
 
 Cadmium 
 
 112 
 
 0567 
 
 6'35 
 
 Carbon . 
 
 12 
 
 
 
 a. Wood charcoa 
 
 
 
 2410 
 
 2-89 
 
 /3. Natural graph te 
 
 
 
 2020 
 
 2-42 
 
 7. Diamond 
 
 
 
 1469 
 
 1-76 
 
 Cobalt 
 
 59 
 
 1067 
 
 6-29 
 
 Copper 
 Gold 
 
 63'3 
 197 
 
 0952 
 0324 
 
 6-02 
 
 6-36 
 
 Iodine 
 
 127 
 
 0541 
 
 6-87 
 
 Iron 
 
 56 
 
 1138 
 
 6-37 
 
 Lead 
 
 207 
 
 0314 
 
 6*50 
 
 Lithium 
 
 7-02 
 
 9408 
 
 6-60 
 
 Magnesium 
 
 24 
 
 2499 
 
 6-00 
 
 Mercury (solid) 
 
 200 
 
 0319 
 
 6-38 
 
 Nickel . 
 
 58 
 
 1092 
 
 6'33 
 
 Phosphorus 
 
 31 
 
 
 
 
 a. Common 
 
 
 I8 95 
 
 5^87 
 
 |8. Red 
 
 
 
 1698 
 
 5-26 
 
 Platinum 
 
 195 
 
 0324 
 
 6-31 
 
Specific Heat and Atomic Weight 115 
 
 Name of the Element 
 
 Atomic 
 
 Weight 
 
 Sp. Ht. of 
 Equal Weights 
 
 Sp. Ht. of 
 Atomic Weights 
 
 Potassium 
 
 39'I 
 
 1655 
 
 6-47 
 
 Silicon . 
 
 28 
 
 
 
 
 a. Graphitoidal 
 
 
 
 1810 
 
 5-07 
 
 j8. Crystallised 
 
 
 
 1650 
 
 4-62 
 
 7. Fused 
 
 
 
 1380 
 
 3-86 
 
 Silver . 
 
 108 
 
 0570 
 
 6-16 
 
 Sodium . 
 
 23 
 
 2934 
 
 6-75 
 
 Sulphur (octahedra ) 
 
 32 
 
 1776 
 
 5-68 
 
 Tin 
 
 118 
 
 0562 
 
 6-63 
 
 Zinc 
 
 65 
 
 0956 
 
 6-23 
 
 The greater number of the specific heats given in this 
 table were determined by Regnault. A glance down the 
 fourth column will show that, with three exceptions (boron, 
 carbon, and silicon), the amount of heat required to pro- 
 duce the same change of temperature in the different 
 elements is nearly the same in all cases when the quantities 
 operated upon are in the proportion of their atomic weights. 
 That the numbers representing the atomic heats are not 
 found to be exactly identical is due partly to unavoidable 
 errors in the estimation of the specific heats, and partly to 
 the fact that the different elements are not dealt with under 
 conditions which are strictly comparable with one another. 
 Thus, solid mercury and solid bromine, at the temperatures 
 at which the specific heats were determined, are much nearer 
 to their melting-points than are the solids, copper and iron, 
 at the temperatures at which the same operation was per- 
 formed upon them. Other circumstances, such as the 
 assumption of different allotropic forms by some of the 
 elements, tend to the introduction of further uncertainty. 
 
 Experiments made by F. Weber in 1876 prove that the 
 specific heats of the three exceptional non-metallic elements, 
 carbon, silicon, and boron, increase rapidly with the tempera- 
 ture, and become nearly constant at high temperatures. At 
 600, or a little above, the number found for diamond was 
 4589, and that for graphite '4670. These numbers multiplied 
 
 I 2 
 
Ii6 Chemical Philosophy. 
 
 by 12 give 5*5 and 5*6 respectively as the atomic heat of 
 carbon in these two forms. It will be observed that these 
 numbers are not appreciably lower than those assigned to 
 several other elements of small atomic weights. The law of 
 Dulong and Petit may, therefore, be now regarded as fully 
 established and liable to no exception. 
 
 The determination of specific heats being always attended 
 by many sources of error, whilst the combining proportion 
 can be fixed with a very considerable degree of accuracy, 
 the application of this rule, like the first, consists essentially 
 in enabling us to decide as to what multiple of the com- 
 bining proportion is to be taken as the atomic weight. In 
 doing this we are guided by the formula 
 
 where 6*2 is the average atomic heat of a solid element. 
 
 Suppose, for example, it is found that 29-5 parts of tin 
 are equivalent to i part of hydrogen, and we require to find the 
 atomic weight. The specific heat of tin is "0562, therefore 
 
 At. Wt. = 6 '1 = 110-3. 
 0562 
 
 The atomic weight of tin is, however, not taken to be 
 no'3, but rather such a multiple of 29-5 as comes nearest 
 to that number, and this is found to be 29*5 x 4 or 118. 
 
 III. Law of Isomorphism. If a crystal of common potash 
 alum is immersed in a saturated solution of the purple 
 chrome alum, the purple salt is deposited uniformly over 
 the colourless nucleus, so that the crystal increases in bulk 
 though it undergoes no alteration of form. The resulting 
 crystal may be transferred to a solution of ammonia alum 
 or of iron alum, or manganese alum, and during every fresh 
 immersion it receives a deposit of a different salt upon its 
 surface, the crystalline form, that of the regular octahedron, 
 being throughout preserved. 
 
 If instead of thus causing successive layers of the various 
 
Isomorphism. 117 
 
 alums to be superposed one upon the other, solutions of any 
 two of these salts are mixed together, crystals of the same 
 form are deposited containing the elements of both salts. 
 
 The alums are double sulphates, all containing the same 
 amount of water of crystallisation, and having a composition 
 which may be represented by the general formula 
 
 M'M'"(S0 4 ),+ 
 
 in which M' may be Cs, Rb, K, Na, Am, Tl or Ag, and M'" 
 may be Fe, Mn, Cr or Al. 
 
 If, therefore, any two of these compounds are compared 
 together, as, for example, 
 
 Potash alum KAl(SO 4 ) 2 .i2OH 2 
 
 Soda alum NaAl(SO 4 ) 2 .i2OH 2 , 
 
 it is obvious that atom for atom they have the same consti- 
 tution, but the one contains potassium, the other sodium. 
 This exchange of an atom of one element for an atom of 
 another is in this case effected without producing any altera- 
 tion in the crystalline structure of the resulting salt, and 
 when bodies thus agree in chemical constitution, and in 
 crystalline form, they are said to be isomorphous. 
 
 From the examination of a great many instances of the 
 same kind, chemists have been led to infer that when two 
 bodies, composed of the same or similar elements, crystal- 
 lise in forms belonging to the same crystallographic system, 
 they generally contain the same number of atoms united 
 together in a similar manner. 
 
 This statement must be considered to include cases in 
 which groups of atoms (compound radicles) take the place 
 and perform the part of single elementary atoms. The 
 compounds of ammonia with acids, for example, are isomor- 
 phous with the corresponding salts of potassium, and a 
 constitution is therefore attributed to the ammoniacal salts 
 similar to that of the potassic salts, the symbols NH 4 being 
 the representative of the metal in these compounds. Thus 
 
1 1 8 Chemical Philosophy. 
 
 in the following pairs of compounds there is the most com- 
 plete concordance in chemical characters as well as in 
 crystalline form: 
 
 Cubical. 
 
 AmCl (Am = NH 4 ) KC1. 
 
 Four or Six-sided Prisms (Trimetric). 
 
 Am 2 SO 4 K 2 SO 4 
 
 Octahedral (Regular). 
 
 AmAl(SO 4 ) 2 i2OH 2 KA1(SO 4 ) 2 i2OH 2 
 Am 2 PtCl 6 K 2 PtCl 6 
 
 Although some of the relations between external crys- 
 talline form and chemical constitution are still involved in 
 obscurity, the existence of a great number of well-marked 
 cases of isomorphism is a fact which is familiar to every 
 chemist, and occasionally the application of this principle 
 has led to the settlement of questions relating to atomic 
 weights, regarding which there had been previously more or 
 less of uncertainty. 
 
 For instance, alumina, the only known oxide of alumi- 
 nium, is believed to have the same constitution as ferric 
 oxide, because not only do the oxides themselves agree in 
 crystalline form, but they are capable of replacing each other 
 in their compounds without disturbing the crystalline struc- 
 ture of these bodies. Now, since ferric oxide is universally 
 regarded as a sesquioxide, that is, containing in each mole- 
 cule two atoms of the metal to three atoms of oxygen, 
 alumina is believed to be formed upon the same type, and 
 if the formula Fe 2 O 3 be employed to represent ferric oxide, 
 A1 2 O 3 must be admitted as the formula for alumina. If 
 these considerations have to be applied to the determina- 
 tion of the atomic weight of the metal, we have only to 
 refer to the analysis of alumina to find that 100 parts 
 contain 
 
 53*3 parts of aluminium, 
 and 467 oxygen. 
 
 And since, according to the formula, we have 3 x 1 6 or 48 
 
Isomorphism. 119 
 
 parts of oxygen united with 2X parts of metal, we can 
 easily calculate the value of x (= 27-4), which is the atomic 
 weight. 
 
 Other instances of a like character would readily present 
 themselves upon inquiry. The following is an interesting 
 example : 
 
 It is well known that the crystalline forms of sodium 
 nitrate and calc-spar are nearly identical, and a crystal of 
 calc-spar immersed in a saturated solution of the nitrate will 
 grow by uniform deposition of that salt all over its surface. 
 Arragonite (another form of calcium carbonate) is also 
 found in prisms of the same form as common potassium 
 nitrate. These facts tend to prove that calcium carbonate 
 probably contains the same number of atoms as the nitrates 
 of potassium and sodium, and that its formula should be 
 
 CaC0 3 
 if the others are KNO 3 
 
 and NaNO 3 respectively. 
 
 But this formula cannot be used unless we assume that 
 the atomic weight of calcium is 40, a number which agrees 
 with the value deduced from other considerations. 
 
 IV. But independently of the existence of vaporisable 
 compounds and of any application of Avogadro's hypothesis, 
 the atomic weight may in some cases be determined by 
 appeal to purely chemical considerations. Take the case 
 of oxygen again. We know by analysis that 100 parts of 
 water contain 88 '88 parts of oxygen to irn parts of hydro- 
 gen. These proportions might be expressed by the formula 
 HO (if O = 8), or by H 2 O (if O = 16), or by H 3 O (if O = 
 24). To decide between these several values we note the 
 results of decomposing water by chemical agents. Potassium 
 acting upon water produces, first, caustic potash, which con- 
 tains all the oxygen, together with halftho. hydrogen of the 
 water. By the further action of the metal the second half 
 of the hydrogen is expelled, and potassium oxide is formed. 
 
12O Chemical Philosophy. 
 
 In this decomposition, then, the hydrogen is displaced 
 in two equal portions, and no reaction is known in which 
 the hydrogen of water is divisible into more than two parts. 
 The oxygen, on the other hand, may be displaced by 
 chlorine, but no such division occurs in this case, and no 
 compound is known in which the whole of the hydrogen 
 in water is associated with half the oxygen and a quantity 
 of chlorine equivalent to the other half. There is no com- 
 pound intermediate between hydrogen chloride and water. 
 
 These facts are expressed in the formula H 2 O, where 
 the atomic weight of oxygen is assumed to be 16 (approxi- 
 mately), and the atomic weight of hydrogen is i. The same 
 value for the atomic weight of oxygen must be admitted in 
 the formulae for ether and for the anhydrides of monobasic 
 acids, like acetic acid. According to views prevailing up to 
 about 1860, alcohol was hydrated oxide of ethyl, EtO.HO, 1 
 and acetic acid was hydrated oxide of acetyl, AcO.HO 2 
 (O being = 8). But the conversion of alcohol into ether, 
 and acetic acid into its anhydride, cannot be readily effected 
 by withdrawing the elements of water. 
 
 In the operations by which these transformations are 
 effected an intermediate compound is always formed, which 
 represents a second molecule of the alcohol or acid. Thus, 
 ether may be produced from alcohol by first displacing 
 from it an atom of hydrogen by potassium, as in the case of 
 water, producing the compound EtOKO, and this is con- 
 verted into ether by making it react with the chloride or 
 iodide of ethyl, EtCl or EtI. So that the resulting ether is 
 EtOEtO. 
 
 The series of compounds thus produced is best repre- 
 sented as parallel with the series formed from water, adopt- 
 ing in both cases and for similar reasons the assumption 
 that each molecule contains one atom and not two atoms 
 of oxygen. 
 
 1 Et = C 4 H 5 , if C is 6. 
 
 2 Ac = C 4 H 3 O2, if C is 6, and O is 8. 
 
Atomic Weights of Oxygen and Carbon. 121 
 
 Water, HHO. Alcohol, EtHO. 
 
 Potassium hydroxide, KHO. Potassium ethoxide, KEtO. 
 Potassium oxide, K 2 O. Ethyl oxide, Et 2 O. 
 
 Again, acetic acid can be deprived of the elements ot 
 water only with the greatest difficulty, even by the action 
 of so powerful a dehydrating agent as phosphorus pent- 
 oxide ; but the anhydride, formerly called anhydrous acetic 
 acid, is instantly produced when acetyl chloride is mixed 
 with acetic acid or a dry acetate. So that the constitution 
 of these compounds is similar to that of the ethyl series just 
 described, thus : 
 
 Acetyl chloride, AcCl. 
 
 Acetic acid, AcHO. 
 
 Potassium acetate, KAcO. 
 
 Acetic anhydride, Ac. 2 O. 
 
 In all these cases the value for the atomic weight of oxygen 
 is necessarily 16 (approx.). 
 
 The valuation of the atomic weight of carbon may be 
 effected in a somewhat similar manner. Marsh gas is the 
 simplest compound of carbon and hydrogen known. It 
 contains 3 parts by weight of carbon to i part of hydrogen, 
 and assuming the value 3 for the symbol C, this composition 
 could be expressed by the formula CH. By the same 
 hypothesis the formula of carbon perchloride is CC1. But 
 when chlorine acts upon marsh gas it forms three inter- 
 mediate products ; hence the hydrogen in marsh gas is 
 divisible into four equal parts, and the hydrogen must be 
 present in the form of four atoms, and the formula becomes 
 C 4 H 4 . If there are four atoms of carbon in a molecule of 
 marsh gas, we must believe that these four remain united 
 together to form a group which remains intact throughout 
 all the vast array of compounds derived from marsh gas, 
 and it would be difficult to assign a reason for the three 
 intermediate stages in the chlorination of marsh gas. So 
 that, since there is no evidence of the divisibility of the 
 carbon in the two oxides of carbon, in marsh gas, in chloro- 
 
122 Chemical Philosophy. 
 
 form, and other allied substances, one molecule of each of 
 these compounds is represented as containing one atom of 
 carbon, to which the proportional value 12 is given. 
 
 It is evident, therefore, that the valuation of atomic 
 weights by chemical methods also involves the determination 
 of the relative molecular weights of a number of compounds. 
 A more direct consideration of this part of the subject is 
 contained in the next chapter. 
 
 CHAPTER XV. 
 
 MOLECULAR WEIGHTS AND FORMULA. 
 
 WHEN a compound has been analysed, it is usual in the first 
 instance to represent its composition by the percentages of 
 the several elements of which it is made up. Acetic acid, 
 for example, contains 
 
 Carbon 40*0 
 
 Hydrogen 6*6 
 
 Oxygen 53-4 
 
 in 100 parts. The next step is to endeavour to write a 
 formula which, whilst expressing the same facts more com- 
 pactly, gives, at the same time, the number of atoms of the 
 constituent elements, and fixes the relative weight of the 
 molecule. 
 
 If the atomic weights of the elements were all equal, the 
 formula would be a mere repetition of the percentages ; but 
 since they are different, the number of atoms of the several 
 elements contained in equal weights must be inversely as 
 their atomic weights. 
 
 The simplest rule for deducing the formula of a com- 
 pound from its percentage composition is, therefore, to 
 divide the respective quantities by the atomic weights of 
 the elements. 
 
Molecular Weights. 123 
 
 Thus, if we divide the percentages of carbon, hydrogen, 
 and oxygen in acetic acid by the atomic weights of carbon, 
 hydrogen, and oxygen respectively, we arrive at these re- 
 sults. 1 
 
 ^ ......... =3-3 
 
 12 
 
 =6-6 
 
 16 
 
 So that evidently there are as many atoms of oxygen pre- 
 sent as there are of carbon, and there are twice as many 
 hydrogen atoms. The simplest formula, then, that can be 
 written for acetic acid is CH 2 O ; but whether this is to be 
 taken as representing a molecule of acetic acid or whether 
 the true formula is some multiple of this, such as C 2 H 4 O 2 
 or C 3 H 6 O 3 , remains to be decided by considerations which 
 now require to be examined. 
 
 VAPOUR DENSITY AND MOLECULAR WEIGHT. 
 
 According to the law of Avogadro, equal volumes ot 
 all true gases, irrespective of chemical composition, contain 
 under the same conditions of temperature and pressure 
 the same number of molecules. It follows from this that 
 the weights of equal volumes must be proportional to the 
 weights of the molecules of which the gases are composed. 
 
 This is the principle of the only direct method for ascer- 
 taining the relative weights of these molecules. 
 
 Comparing together, for example, equal measures of 
 hydrogen and hydrochloric acid gases, we find their respec- 
 tive weights represented by the numbers i and 18-25. But 
 the weight of hydrogen contained in hydrochloric acid is 
 exactly half the weight of the same element contained in an 
 equal measure of hydrogen, and if we assume that there is 
 1 See Examples and Exercises, p. 155, No. 25. 
 
124 Chemical Philosophy. 
 
 one atom of hydrogen in a molecule of hydrochloric acid 
 (and by the theory there cannot be less than one atom), we 
 arrive at the conclusion that the molecule of hydrogen con- 
 sists of at least two atoms. But, further, we have every 
 reason to believe that whilst the molecule of hydrochloric 
 acid contains at least one atom of hydrogen, it does not 
 contain more than one atom. When metals act upon 
 hydrochloric acid the hydrogen is expelled all at once, 
 and not in several portions, as in the case of water and 
 ammonia. So that, assuming the atomic weight of hydrogen 
 as i unit of weight, the molecule of hydrochloric acid must 
 weigh 36-5 units. The weight of hydrogen equal in bulk 
 to this, that is to say, one molecule of hydrogen or two unit 
 volumes, must accordingly be 2. 1 
 
 The same conclusions may be established by a slightly 
 different form of the same argument. Equal volumes of 
 hydrogen and chlorine react to form hydrochloric acid equal 
 in volume to the mixture, or twice the volume of the 
 hydrogen or of the chlorine alone. So that whatever is the 
 number of molecules in the hydrogen gas employed, the 
 matter composing them is distributed into twice the number 
 of molecules of hydrochloric acid. And as there is every 
 reason for supposing that the hydrogen chloride molecules 
 are all exactly alike, each of the original hydrogen molecules 
 are broken up into two equal and similar portions. 
 
 Again, two volumes of hydrogen mixed with one volume 
 
 1 The student will now perceive why 2 volumes and not I volume 
 or 3 or 4 volumes is regarded as the standard volume of molecules. 
 One volume would be inconvenient, because we find that the molecule 
 of hydrogen can be divided into two equal parts, and this would 
 necessitate the use of fractions. Three volumes would be incorrect, 
 because the hydrogen molecule is divisible into two and not into three 
 atoms. Four volumes would also be inadmissible, because the same 
 bulk, that is one molecule of hydrochloric acid, would then be repre- 
 sented as containing two atoms of hydrogen and two atoms of chlorine, 
 H 2 C1 2 . Whereas it is a matter of fact that neither the hydrogen nor 
 the chlorine in hydrochloric acid is divisible into separate parts. 
 
Molecules of Elements. 125 
 
 of oxygen, at any temperature above the boiling-point of 
 water, make three volumes of the mixed gas. If a spark 
 be now passed into the mixture the gases ignite, and on 
 cooling to the original temperature leave two volumes of 
 steam that is, a quantity double the bulk of the oxygen 
 used. Here again the matter composing the oxygen mole- 
 cules is distributed into twice as many molecules of steam, 
 and hence each of the oxygen molecules must have been 
 divisible into two equal and similar parts. 
 
 Regarding the diatomic character of hydrogen and oxygen 
 molecules, other evidence, independent of any hypothesis 
 concerning the physical constitution of gases, is supplied by 
 the following facts. 
 
 Metallic copper is capable of expelling hydrogen from 
 hydrochloric acid only very slowly, even when boiled with 
 it. But by adding hypophosphorous acid to a warm solu- 
 tion of sulphate of copper, a brown precipitate of cuprous 
 hydride is thrown down, and this compound, in contact with 
 hydrochloric acid, evolves hydrogen freely and forms cuprous 
 chloride. This reaction can only be explained upon the 
 assumption that it is the attraction of the hydrogen in the 
 copper hydride for the hydrogen in the acid, superadded to 
 that of the copper for the chlorine, which determines the 
 metathesis : 
 
 rn /H H - Cl 
 
 CU2 \H H - Cl 
 
 rn /Cl H - H 
 
 CU2 \C1 H - H 
 
 A great many reactions of a similar character are known, 
 chiefly among compounds containing a relatively large pro- 
 portion of oxygen, part of which escapes in the gaseous 
 form. Thus, when silver oxide is placed in contact with 
 peroxide of hydrogen, the silver is reduced to the metallic 
 state, water is formed, and oxygen gas evolved. 
 
 Ag 2 O + H 2 O 2 = Ag 2 + H 2 O + O 2 
 
126 Chemical Philosophy. 
 
 In like manner, permanganic and chromic acids are 
 decomposed by peroxide of hydrogen with evolution ol 
 oxygen gas ; and in these and similar reactions it has been 
 proved experimentally that half the oxygen comes from the 
 peroxide, half from the acid or other body with which 
 it is in contact. The conclusion seems inevitable that 
 these two halves of the oxygen had an attraction for each 
 other which was sufficient to upset the equilibrium of the 
 unstable compounds of which they previously formed a 
 part. 
 
 It is also well known that substances in the nascent 
 state that is, at the instant of their liberation from com- 
 pounds are capable of acting far more energetically than 
 when they are employed in the bodily form. Hydrogen,^, 
 for example, is generally incapable of combining with other 
 bodies ; but when materials such as zinc and dilute sul- 
 phuric acid, which are capable of yielding hydrogen, are 
 employed, many decompositions and combinations may be 
 brought about which would be otherwise impossible. Solu- 
 tion of sulphur dioxide may in this way be converted into 
 hydrogen sulphide and water by the action of zinc and 
 diluted hydrochloric acid, the nascent hydrogen attacking 
 and combining with both the sulphur and the oxygen. An 
 instructive experiment, illustrating the power of both nascent 
 oxygen and hydrogen, consists in electrolysing a solution of 
 hydrochloric acid coloured with indigo. The liquid in the 
 neighbourhood of both poles is bleached ; at the negative, 
 because the hydrogen there liberated combines with the 
 indigo and forms a colourless compound ; at the positive, 
 because the chlorine, acting on the water, disengages 
 oxygen, which, whilst still nascent, combines with the 
 elements of the indigo, producing a pale yellow substance. 
 Neither oxygen nor hydrogen in the ordinary gaseous form 
 is capable of producing these effects, which are usually 
 supposed to be due to the superior powers of the atoms 
 whilst still in the free state and before they have partly 
 
Molecular Formula. 127 
 
 expended their energies by coupling in pairs or otherwise 
 binding together in molecules. 
 
 In order, therefore, to express the molecular weights of 
 gases, we must double the numbers representing their 
 relative densities ; and this rule is applied to all gaseous and 
 volatile bodies, with the few exceptions referred to in the 
 next chapter. 
 
 The application of this principle to those of the elements 
 that are volatile leads to the conclusion that many of them, 
 like hydrogen and oxygen, consist of molecules having a 
 duplex structure, though examples are not wanting of a more 
 complex as well as of a simpler constitution. We come, in 
 fact, to the following classification : 
 Monatomic Molecules : Molecular 
 
 Relative 
 
 Density Weight Formula 
 
 Mercury . . . 100 . 200 . Hg 
 
 Cadmium . . . 56 . 112 . Cd 
 
 Zinc .... 32*5 . 65 . Zn 
 
 Iodine (above 1600) . 63*5 . 127 .1 
 Diatomic Molecules : 
 
 Hydrogen ... i 2 . H 2 
 
 Oxygen . . . 16 32 . O 2 
 
 Nitrogen . . . 14 28 . N 2 
 
 Chlorine . 35 '5 71 . C1 2 
 
 Bromine . . . So . 160 . Br 2 
 
 Iodine (below 500) . 127 . 254 . I 2 
 
 Potassium . . . 39*1 . 78*2. K 2 
 
 Sulphur (at 900) . . 32 64 . S 2 
 
 Triatomic Molecule : 
 
 Ozone . . . . 24 48 . O 3 
 
 Tetr atomic Molecules : 
 
 Phosphorus . . . 62 . 124 . P 4 
 
 Arsenic . . . 150 . 300 . As 4 
 
 Hexatomic Molecule : 
 
 Sulphur (at 500) . . 96 . 192 . S 6 
 
128 Chemical Philosophy. 
 
 It may be noticed in the table just given that several 
 elements are capable of existing in the gaseous state in the 
 form of molecules of various degrees of complexity. Thus 
 whilst vapour of sulphur at temperatures not much above 
 the boiling-point of the liquid consists of hexatomic mole- 
 cules, at higher temperatures these are broken up into 
 groups each containing only two atoms. Many compounds 
 undergo a corresponding simplification when heated, and 
 generally it may be stated that even in the gaseous state the 
 number of atoms which remain united to form a molecule 
 is less at high than at low temperatures. This will be 
 referred to in greater detail in the chapter on ' Dissociation.' 
 
 SYNTHESIS AND ANALYSIS OF COMPOUNDS. 
 
 The cases that we have hitherto had under discussion 
 have been chiefly those of the elements. If we have to 
 examine into the constitution of compounds, we have yet 
 several sources of information which will help in the solution 
 of the problem of their molecular weights and formulae. 
 One very important kind of argument is deduced from a 
 knowledge of the mode or modes in which the compound 
 may be formed, and of the products of its decomposition 
 under the influence of reagents. 
 
 For example, oxalate of sodium is formed when carbonic 
 anhydride gas is passed over heated sodium. The change, 
 however, might be represented by either of the following 
 equations : 
 
 Na + CO 2 = NaCO 2 
 or, Na 2 + 2CO 2 = Na 2 C 2 O 4 
 
 But the doubt, if there were any, would be resolved in 
 favour of the latter alternative, by the observation that when 
 sodium oxalate is heated it yields carbonic oxide gas, leaving 
 a residue of carbonate. 
 
 CO 
 Na 2 C0 3 
 
Synthesis of Compounds. 1 29 
 
 The most direct and rational explanation of this is, that 
 a molecule of the salt contains two atoms of carbon, as 
 represented by the formula. 
 
 Again, succinic acid may be built up from its elements 
 by a succession of processes, which are briefly represented 
 in the following series of equations ; and from them we learn, 
 even if no other evidence were forthcoming, that succinic 
 acid contains at least four atoms of carbon in a molecule, 
 and hence must have the formula here assigned to it. 
 
 Carbon poles ignited by the electric current in hydrogen 
 gas give rise to acetylene. 
 
 C/2 "I" Ho ^ C^H^ 
 
 Acetylene can be made to combine with hydrogen, 
 yielding ethylene, the known density of which gas renders 
 impossible any doubt as to its molecular weight. 
 
 C2H 2 + H 2 = 
 
 Ethylene combines with bromine thus : 
 C2H 4 -f- Br2 = C 2 H 4 Br 2 
 
 This dibromide may be converted into a cyanide : 
 C 2 H 4 Br 2 + 2KCN = C 2 H 4 C 2 N 2 + 2KBr. 
 
 Lastly, this cyanide, under the influence of boiling alkali, 
 assimilates the elements of water, and yields up its nitrogen 
 in the form of ammonia. 
 
 Cyanide of ethylene. Potassium succinate. 
 
 C 2 H 4 |gj + 2KHO + 2 H 2 = C 2 H 4 |^| + 2NHg 
 
 From this salt the acid may be procured by the action 
 of sulphuric acid. 
 
 C 2 H 4 (ggiK + 
 
 Potassium Sulphuric Succinic Potassium 
 
 succinate. acid. acid. sulphate. 
 
 K 
 
130 Chemical Philosophy. 
 
 Evidence is also derivable from a study of products of 
 decomposition. Analysis of glucose, for example, gives 
 proportions of carbon, hydrogen, and oxygen, which may be 
 expressed by (CH 2 O) n , a formula identical with that of 
 acetic acid, a totally different substance. The problem is 
 to determine the value of n. When glucose is gently 
 oxidised it yields saccharic acid, a well-established dibasic 
 acid, the two potassium salts of which are respectively re- 
 presented by the formulae KC 6 H<,O 8 and K 2 C 6 H 8 O 8 . 
 Ammonium saccharate when heated splits up according to 
 the following equation : 
 
 (NH 4 ) 2 C 6 H 8 8 = C 4 H 5 N + 2 C0 2 + NH 3 + 4 H 2 O. 
 
 Ammonium saccharate. Pyrroline. 
 
 Further oxidation of saccharic acid yields oxalic acid, 
 H 2 C 2 O 4 , and oxalic acid in turn yields carbonic acid. 
 
 Hence the molecule of saccharic acid contains C 6 , and 
 consequently the molecule of glucose from which it is 
 formed very probably contains at least C 6 H 12 O 6 , inasmuch 
 as products of oxidation almost always contain either the 
 same number of carbon atoms as the compounds from which 
 they were derived, or a smaller number. 
 
 This rule, however, though a very useful guide, is not 
 without exceptions. Thus 
 
 (aldehyd, C 2 H 4 O 
 Alcohol, C 2 H 6 O . yields j and 
 
 (acetic acid, C 2 H 4 O 2 
 
 propionic aldehyd, C 3 H 6 O 
 a propyl alcohol, C 3 H 8 O or 
 
 propionic acid, C 3 H 6 O 2 
 
 (acetone, C 3 H 6 O 
 
 /? propylic alcohol, C 3 H 8 jor acetic acid, C 2 H 4 O 2 
 
 (and formic acid, CH 2 O 2 
 
 Toluene, C 7 H 8 . benzoic acid, C 7 H 6 O 2 
 
Saturating Power of Acids and Bases. 131 
 
 (toluic acid, C 8 H 8 O 2 
 Xylene, C 8 H 10 . . yields j or 
 
 (phthalicacid, C 8 H 6 O 4 
 
 &c. &c. 
 
 But 
 
 Benzene, C 6 H 6 . yields benzoic acid, C 7 H 6 O 2 
 
 Phenol, C 6 H 6 O . first phenoquinone, 
 
 In such cases the condensation results from several 
 molecules being simultaneously attacked. 
 
 The compound last named, phenoquinone, is known to 
 possess the formula ascribed to it because it is otherwise 
 produced by the union of one molecule of quinone, 
 C 6 H 4 O 2 , with two molecules of phenol, C 6 H 6 O, no other 
 product being formed at the same time. 
 
 The determination of molecular weight is, however, 
 more conveniently and directly accomplished either by 
 appeal to vapour density, or, if that is not possible, to one 
 or other of the methods following. 
 
 SATURATING POWER OF ACIDS AND BASES. 
 
 In the examination of acids and bases, and other bodies 
 which are capable of entering into combination readily, or 
 of suffering the replacement of some of their elements by 
 simple reactions which do not involve destruction of the 
 molecule, the process for determining the molecular weight 
 is generally easy. Suppose, for instance, it is required to 
 determine the molecular weight of sulphuric acid, it is only 
 necessary to add to it various quantities of potash or soda 
 to discover that there are two, and two only, distinct and 
 definite sulphates of potassium and sodium, and that even 
 double salts are possible, in which the two metals figure 
 side by side, in place of the hydrogen of the acid. So that 
 
 K 2 
 
32 Chemical PhilosopJty. 
 
 sulphuric acid and its salts are representable by such for- 
 
 )so 4 
 
 mulae as the following : 
 
 | 
 
 Na ^ Na 
 
 al 80 ' 
 
 Tartaric acid is a case of similar kind. Its molecular 
 formula cannot be less than C 4 H 6 O 6 , on account of the 
 existence of the double tartrates, which prove that the acid 
 contains two replaceable basic hydrogen atoms. Thus : 
 
 Cream of tartar . . KHC 4 H 4 O 6 
 Rochelle salt . . . KNaC 4 H 4 O 6 + 4H 2 O 
 Emetic tartar . . . K(SbO)C 4 H 4 O 6 + H 2 O 
 
 Basic derivatives of ammonia are dealt with in a similar 
 way. If we assume that the molecule of hydrochloric acid 
 is represented by the symbols HC1 (= 36-5), the problem 
 is to find what weight of base will enter into combination 
 with 36*5 parts of hydrochloric acid so as to produce a 
 neutral compound. In the case of ammonia itself the 
 compound formed with hydrochloric acid is represented by 
 the formula NH 3 HC1, in which NH 3 stands for 17 parts 
 by weight of ammonia. In some cases it is practically 
 more convenient to prepare the double salts which the 
 hydrochloride of ammonia and all similarly constituted 
 compounds form with platinic chloride. Ammonio-chloride 
 ,of platinum has the formula 2(NH 3 .HCl).PtCl 4 , and that 
 quantity of the basic body which, in this compound, is 
 capable of taking the place of ammonia, NH 3 , is generally 
 taken to be its molecular weight. 
 
 This principle is applied especially in the case of those 
 natural alkaloids which contain oxygen and are not vaporis- 
 able without decomposition. Quinine, from cinchona bark, 
 is an example of the kind. The formula may be either 
 C 10 H 12 NO or C 2 oH 24 N 2 O 2 . The latter is chosen chiefly 
 
Substitution Compounds. 133 
 
 because quinine forms two classes of salts, namely, neutral 
 compounds, in which C 20 H 24 N 2 O 2 represents the NH 3 in 
 ammonia salts, and acid compounds, containing half the 
 proportion of base or twice the proportion of acid, as in the 
 following formulae : 
 
 (C 20 H 24 N 2 2 HC1+2H 2 O 
 Neutral . -I (C 20 H 24 N 2 O 2 HCl) 2 PtCl 4 + 3 H 2 O 
 
 ((C 20 H 24 N 2 2 ) 2 H 2 S0 4 + 8H 2 
 
 (C 20 H 24 N. 2 2 + 2HC1 
 
 Acid . J C 20 H 24 N 2 O 2 .2HCl.PtCl 4 + H 2 O 
 
 * (C 20 H 24 N 2 2 H 2 S0 4 
 
 It appears, therefore, that the number of atoms of nitro- 
 gen in the molecule does not determine the saturating power 
 of the base. This is further illustrated by urea, CH 4 N 2 O, 
 which forms monoacid salts by union with HC1, with 
 HNO 3 , or with ^H 2 SO 4 , &c. ; also by such bases as rosa- 
 niline, C 2 oH 19 N 3 , which combines either with HC1, forming 
 the well-known magenta dye, or with 3HC1, forming a brown 
 crystalline salt. 
 
 SUBSTITUTION COMPOUNDS. 
 
 In other cases, especially when the compound under 
 examination is neutral, and incapable of entering into com- 
 bination with other bodies of known molecular weight, the 
 results of * substitution ' afford information in the direction 
 required. To take an instance, the hydrocarbon benzene 
 is a neutral liquid, neither acid nor basic, which forms no 
 compounds whose constitution throws any light on the 
 question of its molecular weight. It is volatile, and its 
 vapour density would tell all that we require to know, 
 namely, that to represent a molecule of it we must use the 
 formula C 6 H 6 ; but even if we had not this evidence to 
 appeal to, the same result would be indicated by the com- 
 position of the products which are formed from it under the 
 agency of chlorine. Acted upon in this way, benzene yields 
 
134 
 
 Chemical Philosophy. 
 
 first a product in which one atom of chlorine is combined 
 with six atoms of carbon and five atoms of hydrogen. Now, 
 since this can be shown to consist of one homogeneous 
 chemical compound, and not a mixture, the carbon it con- 
 tains must be derived from one molecule of benzene, the 
 formula of which therefore becomes C 6 H 6 . This is con- 
 firmed by the succeeding products, the entire series running 
 as follows : 
 
 ^6^6 
 
 C 6 H 5 C1 
 
 C 6 H 4 C1 2 
 
 C 6 H 3 C1 3 
 
 C 6 H 2 C1 4 
 
 C 6 HC1 5 
 
 C 6 C1 6 
 
 This removal of hydrogen, and its replacement atom for 
 atom by chlorine or by bromine or iodine, is one of the 
 most remarkable facts in chemistry. For notwithstanding 
 the strong electro-chemical opposition of hydrogen and the 
 halogens, they may be usually exchanged for one another 
 in carbon compounds without altering fundamentally the 
 character of the compound. Acetic acid affords a good 
 instance of this. It yields three chloracetic acids, which like 
 itself are crystallisable, .volatile, monobasic acids. 
 
 Name 
 
 Formula 
 
 Melting Point 
 
 Boiling Point 
 
 Acetic acid 
 Monochloracetic acid 
 Dichloracetic acid . 
 Trichloracetic acid . 
 
 HC 2 H 3 2 
 HC 2 H.,C1O 2 
 HC 2 HC1.A 
 HC;C1 3 O 2 
 
 I 7 
 62 
 
 0? 
 
 53 
 
 II9 
 
 186 
 190 
 195 
 
 One other fact often serves as a useful guide in selecting, 
 for carbon compounds especially, a minimum formula. 
 Among the vaporisable compounds in which carbon is 
 united with hydrogen alone, or with hydrogen and oxygen, 
 no example is known in which two volumes of the vapour 
 
Formula of Metallic Salts. 135 
 
 contain an uneven number of atoms of hydrogen. 1 Conse- 
 quently if the simplest formula deduced from analysis 
 includes such a number it cannot represent the molecular 
 weight, and must be at least doubled. Thus the composition 
 of phthalic acid is expressed exactly by the symbols 
 C 4 H 3 O 2 . But according to the present state of knowledge, 
 a molecule of this kind cannot exist ; the molecular for- 
 mula cannot be less than C 8 H 6 O 4 . Whether this cor- 
 rectly represents the molecule can only be settled by other 
 inquiries, the general nature of which has already been indi- 
 cated. 
 
 But, notwithstanding the multiplicity of rules which serve 
 to guide chemists in the selection of formulae whereby to 
 represent molecules, there still remain a large number of 
 bodies which cannot be dealt with by any method at present 
 known. Hence many of the formulae commonly accepted 
 and employed in chemical works are at best expressions of 
 mere guesses enjoying various degrees of probability. Many 
 difficulties occur, for example, among metallic compounds. 
 The formula CrO 3 is generally used for chromic anhydride, 
 not on account of any direct evidence in favour of it, but 
 because of the existence of a volatile oxychloride, CrO 2 Cl 2 , 
 and the analogy of these two compounds with sulphuric 
 anhydride, SO 3 , and the corresponding oxychloride, SO 2 C1 2 , 
 also on account of the isomorphism of the chrotnates and 
 sulphates. Again, potassium permanganate is sometimes 
 expressed by the formula KMnO 4 , which recalls its isomor- 
 phism with the perchlorate, KC1O 4 , but the double formula 
 K 2 Mn 2 O 8 has something to recommend it as satisfying the 
 law of even numbers. 1 So also doubts exist as to the 
 correct mode of representing salts like ferrous and stannous 
 chlorides, there being great probability that the usual 
 formulae, FeCl 2 and SnCl 2 , should be doubled in order to 
 represent the reacting units or molecules of these com- 
 pounds. 
 
 1 See Chap. XVI. 
 
1 36 Chemical Philosophy. 
 
 CHAPTER XVI. 
 
 TYPES VALENCY. 
 
 BODIES which are capable of entering into chemical 
 reactions in the same manner, giving rise under similar 
 circumstances to new products having similar properties, 
 are said to belong to the same type. The idea that the 
 chemical constitution of all known bodies is modelled upon 
 a certain limited number of types supplies a means of classi- 
 fying them according to their modes of transformation. 
 Thus water is capable, in a variety of ways, of exchanging 
 its oxygen and hydrogen for other elements and groups of 
 elements (compound radicles), and the bodies which result 
 from these exchanges retain, more or less perfectly, the 
 chemical deportment of water, and are said to belong to 
 the water type. Caustic potash, for example, is referred to 
 the water type because, like water, it is capable of exchanging 
 its oxygen for sulphur, and its hydrogen for a metal, for an 
 elementary atom like chlorine, or for a group of atoms, such 
 as ethyl or acetyl. As a memorandum of the correspondence 
 in these transformations, the two series of derivatives may 
 be formulated in a similar manner. Thus : 
 
 H1 Q K1 Q 
 
 H/ H/ 
 
 H 1 Kl c 
 
 H/ b H/ b 
 
 H to K lo 
 
 Cl / U Cl / L 
 
 CJU C 2 H> 
 
 H "I A K "l n 
 
 C 2 H 3 0/ C C 2 H 3 0/ C 
 
 &c. &c. 
 
 Formulae of this kind are often spoken of as rational 
 
Rational Formula. 137 
 
 formulae. But water is not the only type. Hydrogen, 
 hydrochloric acid, ammonia, and marsh-gas and other bodies 
 are often referred to as types of decomposition, each of 
 which is imitated, more or less closely, by a considerable 
 number of elements and compounds. The student will 
 readily perceive that the list of bodies which might, for 
 special purposes, be selected as types may be extended and 
 varied almost indefinitely. 
 
 This plan of registering in the formula some of the facts 
 which have been observed as to the possible transformations 
 of a body has led chemists to infer that it is possible to 
 represent symbolically the relative positions of atoms in 
 space. The student will do well to approach such an idea 
 with caution. Rational or descriptive formulae of various 
 kinds are not only valuable as memoranda of possible 
 modes of formation and decomposition, but are positively 
 necessary to enable us to distinguish from one another 
 bodies having the same ultimate composition, but different 
 properties. Constitutional formulae, on the other hand, are 
 very often used with the object of representing the supposed 
 arrangement of the constituents within a molecule. But, 
 although a considerable body of evidence has been accumu- 
 lated, especially in regard to the numerous compounds of 
 carbon, which almost seems to justify the idea, it should 
 not be forgotten that very little is still known regarding the 
 essential nature of molecules and of their constituent atoms. 
 
 Sulphuric acid furnishes a very good example of the kind 
 of fact and argument upon which rational formulae are based. 
 In this case mere analysis tells us only that the compound 
 contains an atom of sulphur, two atoms of hydrogen, and 
 four atoms of oxygen, or SH 2 O 4 . But on examination of its 
 salts we find that both the atoms of hydrogen are replace- 
 able by metals ; and to indicate this basic function of the 
 hydrogen it is the custom to write it at the beginning of 
 the formula, thus : 
 
 H 2 S0 4 . 
 
1 3 8 Chemical Philosophy. 
 
 When sulphuric acid is distilled with phosphoric chloride, 
 it yields two products having respectively the formulae 
 
 SO 3 HC1 and SO 2 C1 2 . 
 
 The former of these bodies results from the removal of an 
 atom of oxygen and an atom of hydrogen from the sulphuric 
 acid, whilst an atom of chlorine is taken up. 
 
 SO 4 H 2 -OH + C1 =SO 3 HC1. 
 
 And this exchange is repeated when the second derivative 
 is produced. 
 
 SO 3 HC1-OH+C1 = SO 2 C1 2 . 
 
 Either of the new compounds will reproduce sulphuric acid 
 when dissolved in water. It seems, therefore, that sulphuric 
 acid is capable of breaking up into the groups SO 2 and 2HO. 
 And this is confirmed by the fact of the production of sul- 
 phates by the union of sulphur dioxide with peroxides, as 
 in these instances : 
 
 PbO 2 + SO 2 = PbO 2 SO 2 
 
 Lead peroxide. Sulphur dioxide. Lead sulphate. 
 
 (HO) 2 + SO 2 = (HO) 2 SO 2 
 
 Hydrogen Sulphur Hydrogen sulphate, 
 
 peroxide. dioxide. or sulphuric acid. 
 
 Such reactions as these and many others are recalled 
 when we write the formula 
 
 or 
 
 H 
 
 in which it may be regarded as a derivative from two 
 molecules of water, each of which has lost an atom of 
 hydrogen, so that the two residues are united together by 
 the group SO 2 . 
 
 And there are many cases in which we write such descrip- 
 
Valency or Combining Capacity. 139 
 
 tive formulae with the utmost confidence that they express 
 possible reactions, although such reactions, for various 
 reasons, may never have been observed. 
 
 VALENCY. 
 
 The formulae that we now make use of are, to a great 
 extent, based upon certain assumptions regarding those 
 chemical properties of atoms which are referred to under 
 the name ' valency.' 
 
 In the formulae 
 
 C1H OH 2 NH 3 CH 4 
 
 we see one atom of chlorine combined with one atom of 
 hydrogen ; one atom of oxygen with two of hydrogen ; one 
 atom of nitrogen with three of hydrogen ; and one atom of 
 carbon with four atoms of hydrogen ; and no compound is 
 known in which one atom of either of these elements 
 chlorine, oxygen, nitrogen, or carbon is united with a 
 larger quantity of hydrogen than is represented here. 
 
 This difference of combining capacity is further illustrated 
 by the fact that when chlorine is made to act upon water, 
 ammonia, or marsh-gas, the hydrogen contained in one 
 molecule of each of these compounds is distributed into so 
 many separate molecules of hydrochloric acid. Thus : 
 
 OH 2 + C1 2 = O + 2HC1 
 
 i molecule 2 molecules of 
 
 of water. hydrochloric acid. 
 
 Again : 
 
 NH 3 + 3 C1 = N + 3 HC1 
 
 i molecule 3 molecules of 
 
 of ammonia. hydrochloric acid. 
 
 The following succession of changes indicates the same 
 thing in the case of marsh-gas : 
 
 1. CH 4 + C1 2 = CH 3 C1 + HC1 
 
 2. CH 3 C1 + C1 2 = CH 2 C1 2 + HC1 
 
 3. CH 2 C1 2 + C1 2 = CHC1 3 + HC1 
 
 4. CHC1 3 + C1 2 = CC1 4 + HC1 
 Thus the four atoms of hydrogen which, by the carbon 
 
140 
 
 Chemical Philosophy. 
 
 in the marsh-gas, were united together into one molecule, 
 measuring 2 volumes, are separated into four molecules or 
 eight volumes of hydrochloric acid. 
 
 c 
 
 H 4 
 
 yields 
 
 In these cases, therefore, the combining value of an atom 
 of chlorine is equal to that of one atom of hydrogen ; it is 
 univalent. We summarise these relations when we say that 
 chlorine is a monad, because we never find it linked in any 
 gasifiable compound to more than one atom at a time. 
 Hydrogen must also be a monad, for in the production of 
 the compound HC1, whatever is the attraction of the chlorine 
 for the hydrogen, there must be an equal attraction on the 
 part of the hydrogen for the chlorine. 
 
 As the result of similar observations, we find that the 
 atom of oxygen is usually diad, of nitrogen triad or pentad, 
 of carbon tetrad, the combining capacity or * valency ' being 
 in each case measured by the number of univalent atoms, 
 such as hydrogen or chlorine, which one atom of these 
 elements can respectively combine with or replace. 
 
 The term 'atomicity' is often used to indicate the 
 greatest number of atoms of one kind or another with which 
 a given atom is ever observed to be united. But combining 
 capacity is so much a matter of circumstances that any word 
 
Combining Capacity is Variable. 141 
 
 which seems to indicate a permanent endowment of the 
 atom, whether as to its form, or motion, or electrical pro- 
 perties, must be objectionable. Many facts seem to point 
 to the conclusion that there is no absolute measure of 
 valency. The capacity of saturation of a given atom de- 
 pends upon the nature of the elements with which it is 
 associated. An atom of sulphur can take up no more than 
 two atoms of hydrogen, but it is capable of forming a com- 
 pound with four atoms of chlorine, or with three atoms of 
 oxygen. In like manner, phosphorus forms the compounds 
 PC1 3 and PC1 5 , but its affinity for hydrogen extends only to 
 three atoms, PH 3 , though a fourth may be taken up if ac- 
 companied by an atom of iodine PH 4 I. It has also been 
 observed that the chlorides corresponding with the highest 
 oxides of many of the metals have not yet been produced, 
 and seem to be incapable of existing. Thus there are the 
 oxides CrO 3 , UO 3 , As 2 O 5 , Ni 2 O 3 , but the chlorides Cr 2 Cl 6 , 
 UC1 5 , AsCl 3 , and NiCl 2 indicate the limits of the capacity 
 of these metals for chlorine. It is interesting to notice that 
 in some cases in which the chloride is missing, the corre- 
 sponding fluoride is known. For example, the fluorides 
 CrF 6 and AsF 5 , the representatives of the unknown chlorides 
 CrCl 6 and AsCl 5 , have been described. 
 
 Combining capacity or valency is also to a large extent 
 dependent upon temperature, increasing to an unknown 
 maximum with fall of temperature, diminishing with rise of 
 temperature to a minimum which is reached only when the 
 substance becomes a perfect gas. For example, HC1 and 
 NH 3 are both gases which, at 500, have no apparent action 
 upon each other. On cooling they gradually unite to form 
 solid sal-ammoniac. But even then the affinities of one or 
 more of the elements present are not exhausted, for ammo- 
 nium chloride unites with other chlorides, forming such 
 compounds as the following : 
 
 2NH 4 Cl.ZnCl 2 
 
 2NH 4 Cl.SnCl 4 
 
 2 NH 4 Cl.PtCl 4 &c. 
 
142 Chemical Philosophy. 
 
 These compounds and sal-ammoniac itself can only exist 
 in the solid state, or perhaps dissolved in water. When 
 heated they yield ammonia and hydrochloric acid gases and 
 other products. The formation of many compounds of this 
 kind leads to the suspicion that the valency of chlorine pro- 
 bably forms the link which binds the other atoms together, 
 and that in the solid state its valency is greater than in its 
 gaseous compounds. The other halogens behave in a similar 
 manner. 
 
 Oxygen, again, is apparently diad or bivalent in water and 
 nearly all other vaporisable compounds containing this ele- 
 ment. But in many solid compounds, especially in salts 
 containing water of crystallisation, the facts can best be 
 explained on the hypothesis of a greater combining capacity, 
 probably amounting to four units or more. 
 
 The variation of valency under varying conditions is 
 analogous to the variation in the number and disposition of 
 the faces of crystals. Every crystallisable substance exhibits 
 a primary or simple Form, made up usually of a minimum 
 number of faces and solid angles. Thus common salt 
 habitually crystallises from water in cubes (6 faces), but from 
 a solution containing urea it yields octahedrons (8 faces). 
 Alum usually forms simple regular octahedrons, but occa- 
 sionally produces dodecahedrons (12 faces), or forms inter- 
 mediate between cube or octahedron and dodecahedron. 
 Again, many substances, called on this account dimorphous, 
 are capable of assuming two distinct crystalline forms be- 
 longing to different crystallographic systems, these different 
 forms being produced most commonly by crystallisation at 
 different temperatures. For example, nitre usually crystal- 
 lises in six-sided prisms of the trimetric system, but when 
 deposited from a hot solution it often forms rhombohedrons 
 of the hexagonal system. In all cases of this kind the one 
 form is more stable than the other, and there is a constant-ten- 
 dency for the less stable to pass into the more stable form. 
 
 Combining power, then, is known to be variable. It is, 
 
Usual Valency of Chief Elements. 
 
 143 
 
 however, useful to remember the capacity of combination 
 habitually shown by the elements in their more familiar 
 compounds, and those which have been sufficiently studied 
 admit of classification into the six divisions displayed in the 
 following table : 
 
 VALENCY OF THE PRINCIPAL ELEMENTS. 
 
 NON-METALS. 
 
 Monads. 
 
 Diads. 
 
 Triads. 
 
 Tetrads. 
 
 Pentads. 
 
 Hexads. 
 
 F 
 
 
 
 B 
 
 C 
 
 N 
 
 S 
 
 Cl 
 
 s 
 
 
 
 
 
 
 
 N 
 
 Si 
 
 P 
 
 Se 
 
 Br 
 
 Se 
 
 
 
 
 
 
 
 P 
 
 
 
 Te 
 
 I 
 
 Te 
 
 
 
 
 
 
 
 
 
 
 
 METALS AND METALLOIDS. 
 
 Ag 
 
 Hg 
 
 Al 
 
 Ir Ro 
 
 V 
 
 Os Ru 
 
 
 Cu 
 
 Au 
 
 
 
 
 
 Co Mn 
 
 
 Pt Pd 
 
 As 
 
 W 
 
 
 Ni Fe 
 
 In 
 
 
 
 Mo 
 
 H 
 
 Cd 
 
 Tl 
 
 Cr 
 
 Mn 
 
 Sb 
 
 
 Li 
 
 Na 
 K 
 
 Zn 
 
 Mg 
 Ca 
 
 V 
 
 Fe 
 
 B 
 
 U 
 
 Pb 
 
 Rb 
 
 Sr 
 
 As 
 
 Sn 
 
 
 Cr 
 
 Cs 
 
 Ba 
 
 Sb 
 Bi 
 
 Ti 
 Zr 
 
 Ta 
 
 Nb 
 
 Mn 
 Fe 
 
 
 Pb 
 
 
 Co 
 Ni 
 
 
 
 An arrangement of this kind necessarily involves the 
 
144 Chemical Philosophy. 
 
 separation of many elements which, in properties, are closely 
 allied together, and the association of others which have 
 very little in common. Thus we have to look for thallium 
 among the triads, although it has strong points of resem- 
 blance on the one hand to the alkali metals, and on the 
 other hand to lead. Boron, again, is separated from carbon, 
 aluminium from chromium and iron, lead from silver. 
 
 CONSTITUTIONAL AND GRAPHIC FORMULAE. 
 
 The compound resulting from the union of two or more 
 atoms is called a saturated compound, when the valency 
 of each atom present is satisfied. This condition is ful- 
 filled when one atom of a monad is combined with another 
 monad, or when two atoms of a monad combine with one 
 atom of a diad, or three of a monad with one of a triad, 
 four monad or two diad atoms with one tetrad, and so forth. 
 
 Examples of various orders of compounds are shown in 
 the following formulas, in which the combining power of 
 each atom present is supposed to be neutralised by that of 
 other atoms. In the notation here introduced, it must be 
 understood that the symbol placed on the left of a formula 
 represents an atom to which all on the same line are directly 
 united. Those also which are connected by a bracket are 
 united together. Thus, 
 
 rcH 3 
 
 ICH 3 
 
 means that there are two atoms of tetrad carbon united by 
 one-fourth of their combining power, whilst each retains 
 three atoms of hydrogen. The same relations are expressed 
 in this figure, or ' graphic formula:' 
 
 H H 
 
 H C C H 
 H 
 
Constitutional Formula. 145 
 
 Of course the lines connecting the symbols are not de- 
 signed to represent any substantive bond or link, but merely 
 indicate the manner in which the combining capacity of 
 each atom is disposed of. 
 
 Hydrochloric acid. 
 
 HC1 or 
 
 H Cl 
 
 Water. 
 
 OH 2 or 
 
 H O H 
 
 Copper sulphide. 
 
 CuS or 
 
 Cu=S 
 
 Hydrogen peroxide 
 
 /OH 
 {OH or 
 
 H O O H 
 
 
 Cl 
 
 Phosphoric chloride. 
 
 PC1 5 or 
 
 Cl P-C1 
 Cl Cl 
 
 Aluminic chloride 
 
 A1C1 3 or 
 
 Cl 
 
 Stannic chloride. 
 
 SnCl 4 or 
 
 Cl 
 Cl-Sn Cl 
 
 jri 
 
 Cl 
 O O 
 
 Phosphoric oxide. 
 
 PO 2 or P O P 
 
 . O II II 
 
 lP0 2 O O 
 
 Acetic acid. H O 
 
 ; CH 3 | ii 
 
 \CO(OH) H C C-0 H 
 
 H 
 
146 Chemical Philosophy. 
 
 Sulphuric acid. H (X //O 
 
 S0 2 (HO) 2 or >S^ 
 
 H O/ ^O 
 
 Potassium dichromate. O O 
 
 fCr0 2 (OK) || |j 
 
 ^ O or K O Cr O Cr O K 
 
 [CrO 2 (OK) || (I 
 
 O O 
 
 LAW OF EVEN NUMBERS. 
 
 When several compounds are formed by the union of 
 two elements in different proportions, it is very commonly 
 noticed that the change of valency or combining capacity 
 of the central atom to which the rest may be supposed to 
 be attached takes place by pairs of units. 
 
 Phosphorus, for example, forms two chlorides ; nitrogen 
 combines with three atoms of hydrogen in ammonia, and 
 with four atoms of hydrogen and an atom of chlorine in 
 chloride of ammonium. Sulphur also yields compounds, in 
 which one atom of that element is combined with two atoms 
 of hydrogen, with two atoms of oxygen, and with two atoms 
 of oxygen and two atoms of chlorine. The formulae of 
 these compounds are represented as follows : 
 
 PC1 3 NH 3 SH 2 
 
 PC1 5 NH 4 C1 S0 2 
 
 s ?0 2 
 
 S ici 2 
 
 It will be observed that the difference in the first case 
 amounts to two atoms of chlorine, which represent two 
 atoms of hydrogen, the unit of valency. In the second 
 case, one atom of hydrogen and one atom of chlorine have 
 been added. In the third, the oxide SO 2 may be taken 
 to represent a hypothetical hydride SH 4 , whilst the oxy- 
 chloride corresponds with the unknown compound SH 6 . 
 In each series, the advance in combining power is equivalent 
 
Artiad and Perissad Elements. 147 
 
 to the assumption of two atoms of hydrogen. And so it is 
 in a large number of other cases. 
 
 It seems, therefore, that, as a general rule, the index of 
 valency of any given atom is either an even or an odd 
 number ; or, as it has been expressed, elements are uniformly 
 either * artiad ' or ' perissad.' * And consequently the sum 
 of the indices of valency of all the atoms present is an even 
 number. (See also p. 135.) 
 
 But there are not wanting exceptions to these statements ; 
 and although the number is at present not great, their 
 marked character is sufficient to destroy much of the ap- 
 parent significance of those more numerous instances which 
 conform to the general rule. 
 
 The following are some of the most notable excep- 
 tions: 
 
 Nitric Oxide. This most remarkable compound is a 
 colourless gas, difficult to liquefy, almost insoluble in water } 
 and unchanged by heat. It exhibits all the characteristics 
 of an unsaturated compound. Thus it unites with oxygen, 
 with chlorine, with sulphuric anhydride, and with many 
 metallic salts. 
 
 It is composed of 14 parts of nitrogen with 16 parts of 
 oxygen ; its relative density is 15 (H=i), and consequently 
 its molecular weight is 30. It therefore contains one atom 
 of nitrogen (perissad), combined with one atom of oxygen 
 (artiad), and thus it breaks the law of even numbers. This 
 difficulty might be avoided by employing the double formula, 
 
 N 2 O 2 N=O N-O N=0 N O 
 
 || .| or ||| or |||| | 
 N-O N=0 N-O 
 
 or | or || . or or 
 
 N=0 
 
 but that its low density and difficult condensability point 
 conclusively to the simpler expression NO as the symbol of 
 its molecule. 
 
 Nitric Peroxide. This compound is described in Chapter 
 
 1 &PTIOS, even, and vepiff<r6s t odd. 
 
 L2 
 
148 Chemical Philosophy. 
 
 XVIII. At i5o-2oo the formula NO 2 represents two 
 volumes of the gas ; but even at much lower temperatures 
 this compound exists mixed with the tetroxide, N 2 O 4 . 
 
 Tungsten and Molybdenum Pentachlorides^ WCl*> and 
 MoCl-y These compounds afford very remarkable instances 
 of the association of an artiad atom with an uneven number 
 of perissad atoms, and consequent infraction of the law 
 under discussion. Hexchlorides of both these elements 
 corresponding with their trioxides, exist ; but these chlorides 
 when heated are split up into pentachlorides and free chlorine. 
 The pentachlorides are volatile without decomposition, and 
 the vapours exhibit normal densities. 
 
 COMPOUND RADICLES. 
 
 Any portion of a molecule which is capable of being 
 detached and transferred to some other molecule is called a 
 ' radicle,' whether it consist of a single atom or of a group 
 of atoms. The term 'compound radicle' is, however, not 
 usually applied to a group unless it makes its appearance in 
 several different bodies. Compound radicles present differ- 
 ent degrees of valency, just as do the atoms of which they 
 are built up, so that they are capable of linking together 
 various proportions of other elementary or compound radi- 
 cles. This fact may be experimentally verified by such 
 reactions as the following: 
 
 Ordinary disodic phosphate is alkaline to test-paper; 
 silver nitrate is neutral. When these two salts are mixed 
 together in equivalent proportions a yellow precipitate of 
 phosphate of silver is thrown down, whilst the liquid be- 
 comes strongly acid. The reason of this is apparent when 
 we express the metathesis in the form of an equation. 
 
Compound Radicles. 
 
 149 
 
 AgNO, 
 
 = 
 
 
 Ag") 
 Aglp0 4 
 AgJ 
 
 AgN0 3 
 
 AgN0 3 
 
 NaNO., 
 
 NaN0 3 
 
 One molecule. Three molecules. One molecule Two molecules of 
 
 of silver phosphate. sodium nitrate. 
 
 HN0 3 
 
 One molecule 
 of nitric acid. 
 
 The group PO 4 is trivalent, and so it holds together the 
 two atoms of sodium and one atom of hydrogen in one 
 molecule. But when the interchange occurs, these become 
 respectively united with three (NO 3 ) groups, each of which 
 is univalent, and incapable of connecting itself with more 
 than one atom at a time. Three new molecules result, one 
 of which is nitric acid, the presence of which can be recog- 
 nised by test-papers. 
 
 Although nearly all unsaturated compounds are capable 
 of entering into combination, and many of them perform 
 the part of well-defined radicles, it does not follow that all 
 radicles should be capable of isolation, and the definition 
 of a radicle by no means involves this idea. On the con- 
 trary, we have already examined phenomena (pp. 123 to 126) 
 which indicate that when atoms such as H, Cl, O, or N 
 are liberated from combination, they combine in pairs, and 
 thus satisfy each other's attractions, unless they find them- 
 selves in the presence of other radicles with which they can 
 immediately unite. Compound radicles resemble elementary 
 atoms in this respect. None of the following groups, for 
 example, are known in the free state, the formulae l repre- 
 senting semimolecules, or what may be termed chemical 
 atoms of these radicles. 
 
 1 The dashes serve to indicate the usual valency of each group. 
 
1 50 Chemical Philosophy. 
 
 Hydroxyl (OH)' contained in acids, alcohols, and metallic 
 
 hydroxides. 
 
 Potassoxyl (OK)' contained in potassium oxysalts. 
 Cyanogen (CN)' contained in cyanides. 
 Ammonium (NH 4 )' contained in the salts of ammonia. 
 
 Arsendimethylf As^TT 3 ) in kakodyl and its compounds. 
 V */ 
 
 Methyl (CH 3 y in methylic alcohol and derivatives. 
 Amidogen (NH 2 )' in primary amines and amides. 
 Methenyl (CH)'" in chloroform and similar bodies. 
 
 When displaced from any of their compounds they do 
 not remain isolated, but unite in pairs, producing molecules 
 which in some cases are stable enough to maintain an 
 independent existence. 
 
 We have, for example 
 
 Radicle Name. Corporate Name. 
 
 Hydroxyl . Hydric Peroxide (OH) 2 or O 2 H 2 
 Potassoxyl . Potassic Dioxide (OK) 2 or O 2 K 2 
 Cyanogen . Cyanogen Gas (CN) 2 or C 2 N 2 
 
 Arsendimethyl Kakodyl As^R 3 or As 2(CH 3 ) 4 
 Methyl . Ethane (CH 3 ) 2 or C 2 H 6 
 
 Amidogen {*$& <NH 2 ) 2 or N 2 H 4 
 Methenyl . Acetylene (CH) 2 or C 2 H 2 
 
 The only free monad radicles known are the two bodies 
 already described, namely, nitrosyl, or nitric oxide, NO, 
 and nitryl, or nitric peroxide, NO 2 . Instances of free diad 
 radicles are, however, more numerous. Thus we have 
 
 Mercury . . . . . Hg" 
 
 Cadmium .... Cd" 
 
 Carbonic oxide (carbonyl) . (CO)" 
 
 Sulphur dioxide (thionyl) . . (SO 2 )" 
 
 Ethylene . . . (C 2 H 4 )" 
 
 Ammonia .... (NH 3 y 
 
Salt Radicles. 151 
 
 Molecules of this kind are of the same order as those re- 
 ferred to at the beginning of the chapter; but why in so many 
 cases the number of unemployed units of valency should be 
 an even number, has not yet been satisfactorily explained. 
 
 Accepting the definition of a * radicle' given on p. 148, 
 it is obvious that there are a great many commonly recog- 
 nised radicles which can hardly be expected ever to assume 
 a bodily existence apart from the compounds in which they * 
 occur associated with elements of a different chemical 
 character. 
 
 In all the carbonates, for example, a group consisting of 
 one atom of carbon and three atoms of oxygen occurs, and 
 this group is capable of being exchanged for C1 2 or (OH) 2 
 or O by double decomposition. It is, therefore, entitled 
 to be spoken of as a compound radicle, although, by reason 
 of the large proportion of oxygen it contains, its condition 
 would be that of unstable equilibrium, even if it could 
 assume temporarily an isolated existence. 
 
 Similar remarks apply to such radicles as NO 3 (of 
 nitrates), C1O 3 (of chlorates), SO 4 (of sulphates), PO 4 (of 
 phosphates), and the rest, which, under possible experi- 
 mental conditions, have never yet been isolated. 
 
 But, after all, it is necessary to remind the student that 
 our system of notation is to be understood and employed 
 only in a unitary sense. Every molecule must be regarded 
 as one entire and undivided unit, whose actions and re- 
 actions proceed from the resultant of all the different forces 
 exerted by its several constituent parts. A chemical com- 
 pound may be compared to a musical chord, constituted, 
 doubtless, of many and complex elements, but communi- 
 cating to the ear the impression of singleness and harmony. 
 And though there are many cases in which one constituent 
 may have a predominating influence in virtue of its mass or 
 special store of unexhausted energy, it never acts as though 
 it were alone, its behaviour being always modified by the 
 presence of the other elements with which it is associated. 
 
152 
 
 Chemical Philosophy. 
 
 The doctrine of radicles, no less than that of valency, 
 and the graphic notation founded upon it, is at present to 
 be regarded solely in the light of a convenient but not 
 absolutely necessary system of recording and comparing 
 facts concerning the changes of composition to which bodies 
 are subject under the influence of chemical attraction. 
 
 Note. The following memoranda will serve to assist the 
 student in writing the formulae of many common salts. In 
 order to construct any required formula it is only necessary to 
 place a symbol or group of symbols, taken from under the 
 positive sign, side by side with a symbol or group taken from 
 under the negative sign, and to adjust the quantity of each so 
 as to comply with their respective habits of combination. 
 Thus, let 
 
 R' represent a univalent radicle, 
 R" bivalent radicle, 
 R"' trivalent radicle, &c. 
 Then it is only necessary to remember that 
 
 R' combines with R', 
 2 R' R", 
 
 orR" R", 
 
 3 R/ R "' 
 
 ~ T>f n "R"' &{* 
 
 3 j\. 2 is. , otc. 
 
 In this way the student will readily learn to compose the 
 unitary formulae of all the most commonly occurring compounds, 
 without risk of falling into any serious error. 
 
 VALENCY OF COMMON SALT RADICLES. 
 
 Univalent. 
 
 + 
 
 Bivalent. 
 - + 
 
 Trivalent. 
 + 
 
 Quadrivalent. 
 
 + 
 
 Sexvalent. 
 + 
 
 F H 
 Cl K 
 
 O Ba 
 
 S Sr 
 
 PO 4 (ortho) As 
 As0 3 Sb 
 
 SiO 4 (ortho) Sn (ic) 
 P 2 O 7 (pyro) Pt 
 
 Cr a (ic) 
 
 Br Na 
 
 SO, Ca 
 
 AsO 4 Bi 
 
 
 Fe a (ic) 
 
 I NH 4 
 
 S0 4 Mg 
 
 B0 3 Au 
 
 
 Mn a (ic) 
 
 CN Ag 
 
 S 2 O 3 Zn 
 
 
 
 
 HO 
 
 C0 3 Cd 
 
 
 
 
 NO a 
 NO, 
 CIO 
 
 C 2 4 Hg (ic) 
 C 4 H 4 O a Cu (ic) 
 SiO,(meta)Pb 
 
 
 
 
 C1O 8 
 CIO, 
 
 CrO 4 Fe (ous) 
 Mn0 4 Mn (ous) 
 
 
 
 
 CIO. 
 
 Mn 2 O 8 Sn 2 (ous) 
 
 
 
 
 P0 3 (meta) 
 C a H,O a 
 
 Hg 2 (ous) 
 Cu a (ous) 
 
 
 
 
Exercises on Section III. 153 
 
 EXERCISES ON SECTION III. 
 
 1. ONE atom of antimony is said to be equivalent to three, and one 
 atom of zinc to two, atoms of sodium. Explain this statement. 
 
 2. Distinguish between atomic, equivalent, and molecular weights. 
 Give the atomic and equivalent weights of mercury, zinc, chlorine, 
 
 iodine, sulphur, iron, and copper. 
 
 Also write down the molecular weights of H 2 S, PC1 5 , AsH 3 , 
 H 2 SO, 
 
 3. What weight of sulphuric acid can be precipitated by one gram 
 of barium chloride ? 
 
 4. What is the weight and volume (at normal temperature and 
 pressure) of the hydrogen contained in 10 grams of microcosmic salt, 
 NaNH 4 HPO 4 -4H 2 O? 
 
 5. Enumerate very briefly the various methods by which atomic 
 weights may be determined ; and indicate in the case of each of the 
 following elements the method or methods which would be applicable: 
 oxygen, chlorine, potassium, mercury, carbon, sulphur, lead, silver, 
 arsenic, silicon, barium, copper, manganese. 
 
 6. The specific heat of iron is "1138. State approximately its 
 atomic weight. 
 
 7. The specific heat of cadmium is '0567, and its equivalent 56. 
 Give its atomic weight. 
 
 8. The equivalent of platinum is 49*4, and its perchloride has the 
 formula PtCl 4 . Find its specific heat. 
 
 9. The formula of water was formerly written thus, HO, and 
 subsequently, for some years, H 2 O 2 (assuming O = 8). Discuss both 
 these formulae, pointing out any inconsistencies you may detect in them. 
 
 10. Complete the equation 
 
 Ag 2 + H 2 2 = 
 
 Quote analogous reactions, and explain the theoretical significance 
 of these facts. 
 
 11. The volume of the molecule of a compound body in the gaseous 
 state is double the volume of the atom of hydrogen. Examine the 
 truth of this statement ; give the experimental facts upon which it is 
 based, and discuss any exceptions to it with which you are acquainted. 
 
 12. Acetic acid contains C 40, H 6'6, and O 53*4 per cent.; and 
 chloracetic acid contains 37-5 per cent, of chlorine. Calculate the 
 molecular weight of acetic acid. 
 
154 Chemical Philosophy. 
 
 13. Explain the signification of the several formulae for potassic 
 sulphate } 
 
 K 2 O.S0 3 ; K01 S 
 KOj 
 
 O 0-K 
 and 
 
 14. What is understood by the terms valency and atomicity re- 
 spectively, and how would you ascertain the valency and atomicity of 
 a given element for example, of carbon or phosphorus ? 
 
 15. What is the atomicity of each of the following radicles : 
 
 S, O, Cl, OH, NH 4 , NH S , NH 2 , NH, N, N 3 PO, SO 2 ? 
 
 16. With the help of the table on p. 152 write down the formulae 
 of the following salts : sodium fluoride, silver sulphate, mercuric 
 cyanide, mercurous phosphate, barium chlorate, bismuth chloride, fer- 
 rous orthosilicate, cupric acetate, ferric nitrate, chromic oxalate, stannic 
 phosphate, calcium hypochlorite, &c. &c. 
 
 17. Write the formulae of nitric oxide, carbonic oxide, white arsenic, 
 acetic acid, benzoic acid, oxalic acid, aluminium chloride, ferric 
 chloride, orthophosphoric acid, and describe briefly the method or 
 methods by which the molecular weight of each has been determined. 
 
 HI f H 
 
 1 8. Taking hydrochloric acid HC1, water TT >O, ammonia N< H 
 
 LH 
 
 and carbon dioxide C < Q as typical, write out the formulae of about 
 
 a dozen substances which conform to these types. 
 
 19. If the specific heat of a metal is -03, state approximately its 
 atomic weight. Calculate the exact value of the atomic weight, the 
 composition of the chloride being, metal 2^915 parts to I part of 
 chlorine by weight. 
 
 20. The atomic weight of silver being 108, and its specific heat "057, 
 another metal M, of which 70 parts unite with 35-5 parts of chlorine, 
 is found to have the specific heat '0306. What is the atomic weight 
 of this metal and the formula of its chloride ? 
 
 21. Give reasons for representing hydrobromic acid by a formula 
 similar to that of hydrochloric acid ; HC1, HBr. 
 
Exercises on Section III. 155 
 
 22. Alcohol, ether, and acetic ether have the following rational 
 formulae : 
 
 C 2 H 5 | Q 
 
 3 0} 
 
 H C 2 H 5 C 2 H 
 
 What arguments could you draw from the existence of such bodies in 
 favour of the number 16 as the atomic weight of oxygen ? Why is it 
 probably neither 8 nor 32 ? 
 
 23. The formula of the molecule or chemical unit of ammonia is 
 NH 3 . What is the meaning of this formula, and what are the reasons 
 for using it ? 
 
 24. A compound is found by analysis to have the following com- 
 position : 
 
 Carbon 52-18 
 
 Hydrogen 13*04 
 
 Oxygen . . . . . . . 3478 
 
 lOO'OO 
 
 To find its simplest formula. 
 
 25. The analysis of a compound leads to these numbers : 
 
 Carbon 37*2O 
 
 Hydrogen 7-90 
 
 Chlorine . . . . . . 54'95 
 
 100-05 
 
 It is not often that the formula can be calculated so easily as in the 
 example given in the text and in the last exercise. It must be borne in 
 mind that in actual practice a slight loss is incurred in the estimation of 
 many elements. The number for hydrogen, however, generally comes 
 out a trifle too high. Oxygen is always estimated by taking the differ- 
 ence between the total weight of the body analysed and the sum of 
 the weights of the constituents which have been actually weighed. 
 In the present example we proceed in the following manner : 
 Divide the percentages by the atomic weights in the usual way : 
 
 2L? = 3-1 atoms of carbon, 
 -j- = 7*9 f hydrogen. 
 
 54_95 _ j .,-4 atom O f chlorine. 
 35'5 
 
1 56 Chemical Philosophy. 
 
 Divide the three quotients by the last, which is the least. 
 - as 2-01 atoms of carbon. 
 
 22. = 5-12 hydrogen. 
 ''54 
 
 Llf _ ! a t om o f chlorine. 
 i'54 
 
 Now, recollecting that the percentages found by analysis are not 
 exactly true, but only close approximations to the correct numbers, and 
 remembering that the hydrogen is generally in slight excess, we may 
 safely reject the two small fractions which occur in the above numbers, 
 and the formula then reads 
 
 C 2 H 5 C1. 
 
 To prove that this represents correctly the composition of the body, 
 it is well to recalculate the percentages on the basis of this formula. 
 This calculation is performed in the following manner : 
 
 C 2 = 12 X 2 = 24 
 
 H 5 = 1x5= 5 
 Cl = 35*5 
 
 64-5 
 
 37- 
 
 55 , 3 
 
 And these theoretical numbers are seen to be very close to those 
 obtained by experiment. 
 
 Theory. Experiment. 
 
 C 37-20 ....... 37-20 
 
 H 775 ....... 7-90 
 
 ci 55-05 ....... 54-95 
 
 26. What is the simplest formula you would assign to a substance 
 containing 
 
 Carbon ....... 54'5 
 
 Hydrogen . ...... 9'2 
 
 Oxygen ....... 36-3 
 
 in 100 parts? 
 
Exercises on Section III. 157 
 
 27. Also to the following body : 
 
 Carbon 88-20 
 
 Hydrogen ...... ii'So 
 
 100-00 
 
 28. And again to an organic base containing 
 
 Carbon 63-78 
 
 Hydrogen. . . . . . 576 
 
 Nitrogen ....... 3-32 
 
 Oxygen 27-14 
 
 100-00 
 
 29. From the following percentages calculate formulae for the 
 several compounds : 
 
 I. 
 
 Iron 60-5^1 
 
 Sulphur 39*5 / 
 
 II. 
 
 Iron 70 *i 
 
 Oxygen 3OJ 
 
 III. 
 
 Hydrogen 5*881 
 
 Oxygen 94-12 J 
 
 IV. 
 
 Sodium 32-79 i 
 
 Aluminium . . . . . 13-02 > 
 
 Fluorine ....... 54-19 J 
 
 V. 
 
 Carbon 39*311 
 
 Hydrogen 7-71 V 
 
 Oxygen 52'98j 
 
 VI. 
 
 Carbon 68-67^ 
 
 Hydrogen 4-95 I 
 
 Oxygen 26'38J 
 
 VII. 
 
 Carbon 49-05 ^ 
 
 Hydrogen 5-14 I 
 
 Nitrogen 28-61 f 
 
 Oxygen i7'2oJ 
 
158 Chemical Philosophy. 
 
 VIII. 
 ... . . . 42-oo-j 
 
 :n 6-46 \ 
 
 VIII. 
 
 Carbon . . . . . . . 42-00- 
 
 Hydrogen 
 
 Oxygen 
 
 IX. 
 
 Carbon 357M 
 
 Hydrogen . . . . . .2-38 
 
 Nitrogen 33-33 
 
 Oxygen 28-58, 
 
 30. Find the formula of nitrosoterpene from these numbers : 
 
 Carbon .*.... 72-57-1 
 
 Hydrogen ..... 8-97 sin 100 parts, 
 
 Nitrogen 8-74] 
 
 and for nitrosoterpene hydrochloride from the following percentages : 
 
 Carbon 59-58 
 
 Hydrogen 8-07 
 
 Nitrogen ....... 7-20 
 
 Chlorine I7'45 
 
 31. The silver salt of an organic acid was found by analysis to yield 
 47-1 per cent, of metallic silver. Determine its molecular weight. 
 
 In the formation of the silver salt from the acid, 108 parts of silver 
 take the place of I part of hydrogen. Therefore, 
 
 Molec. wt. of acid I = molec. wt. of salt 108 
 or Molec. wt. of acid = molec. wt. of salt 1 08 + I. 
 
 In the example given 47-1 parts of silver are contained in 100 
 parts of the salt. So that 1 08 parts of silver are contained in 
 
 108 x iq =229 . 3 parts O f t he salt. This is, therefore, a number 
 
 47-1 
 
 identical with, or very near to, its molecular weight. The answer is, 
 therefore, 
 
 Molec. wt. required = 229-3 108 + I = 122-3. 
 
 Or, since the atomic weights of carbon, hydrogen, oxygen, and 
 nitrogen are all integers, the fraction must be discarded, and the 
 number becomes 122. 
 
 This corresponds with the formula of benzoic acid ; verify it. 
 
 32. Aniline contains 
 
 Carbon 77*4^ 
 
 Hydrogen 7 *S I per cent. 
 
 Nitrogen ....... 15*0 J 
 
Exercises on Section III. 159 
 
 and its platino-chloride contains 32-9 per cent, of platinum : to find its 
 molecular weight and formula. 
 
 As explained in Chapter XV., p. 132, the platinum salts of nitro- 
 genous bases are constituted on the same type as that of ammonia. 
 Hence we may represent the formula of platino-chloride of aniline 
 thus 
 
 2(Aniline + HC1) + PtCl 4 
 
 or Aniline + HC1 + ?*!3i 
 
 2 
 
 The first question, then, is, what weight of platinum salt is repre- 
 sented by this formula ? This is answered as follows : 
 
 32 '9 parts of platinum make 100 parts of platinum salt ; 
 therefore 195 parts or one atom of platinum make 
 
 parts, or 592. 
 
 We have now to subtract from this the platinum perchloride and 
 hydrochloric acid j half the remainder is the molecular weight of the 
 aniline. 
 
 592-337-73=182 
 and 82 
 
 Now, taking the percentage composition of aniline, we have to 
 calculate the proportions of the three elements contained in 91 parts 
 of the base, 
 
 loo : 91 : : 77-4 : x 
 
 x = 70 '4 carbon. 
 100 : 91 : : 7-5 : y 
 
 y = 6 '8 hydrogen. 
 
 loo : 91 : : 15-0 : * 
 
 z = 13*6 nitrogen. 
 
 Hence the formula is obtained by dividing these numbers by the 
 respective atomic weights. 
 
 5 -86 atoms of carbon. 
 
 AQ 
 
 =6*8 atoms of hydrogen. 
 -^ = '97 atom of nitrogen. 
 
160 Chemical Philosophy. 
 
 Hence, allowing for experimental error, which in this case is very 
 small, the molecular formula required is 
 
 C51S6 H 6^_ N -97 or CRN> 
 97 '97 '97 
 
 33. "1442 gram of anthraflavic acid gave '3712 gram of CO 2 and 
 0448 gram of water. Calculate a formula. 
 
 34. A sulphide of tellurium and arsenic was analysed. '6347 gram 
 of the mineral gave '2584 gram of tellurium, "3978 gram of ammonio- 
 magnesium arsenate (MgNH 4 AsO 4 .H 2 O), and 1-6453 g ram of barium 
 sulphate. Calculate a formula. 
 
 35. The analysis of trichloracetyl urea gave the following results : 
 
 (a) -3210 gram gave -2060 gram of CO 2 and '0453 gram of H 2 O ; 
 
 (b) '0825 gram gave '0109 gram of nitrogen; 
 
 (c) '1204 gram gave '2510 gram of AgCl. 
 
 Calculate the formula of the compound. 
 
 36. Analysis of uranium pentachloride : 
 
 Weight of substance taken . . . '8955 gram 
 U 3 O 8 found .... -6038 
 ,,AgCl 1-4997 
 
 Calculate the formula. 
 
 37. -3807 gram of benzoic acid gave '9575 gram of CO 2 and -1698 
 gram of water. And '4287 gram of benzoate of silver gave '2020 of 
 silver. Calculate the rational formula of benzoic acid. 
 
 38. '5828 gram of platino-chloride of caffeine left after ignition -143 
 gram of platinum. What are the molecular weight and formula of 
 caffeine, which contains 
 
 Carbon 49 -05^ 
 
 Hydrogen ...... 5 '14 1 
 
 Nitrogen 2 8-6, P er cent ' ? 
 
 Oxygen 17-20] 
 
 39. The platinum salt of a volatile organic base was found by 
 analysis to have the following percentage composition : Carbon, 9-5 ; 
 hydrogen, 3*2; nitrogen, 5-7; chlorine, 42-0; and platinum, 39-0. 
 The vapour density of the base was found to be 1*59 (air = i). 
 Calculate from these data its molecular formula. 
 
 40. The silver salt of an organic acid contained 62-44 P er cent, of 
 metallic silver. It also contains 17-34 per cent, of carbon and 1-73 
 per cent, of hydrogen. From these data endeavour to find a formula 
 for the acid. 
 
161 
 
 SECTION IV. 
 
 CHAPTER XVII. 
 
 CONDITIONS OF CHEMICAL CHANGE. 
 
 WE now proceed to discuss the general conditions under 
 which chemical changes are brought about, together with 
 some of the circumstances which have been found to 
 modify the exercise of chemical attraction. 
 
 1. Bodies act upon each other chemically only when, 
 according to the usual phrase, they are in absolute contact, 
 that is to say, when they are so near to each other that the 
 distance between them is immeasurably small. 
 
 In this respect chemical attraction differs from the attrac- 
 tion of gravitation, and of electrical and magnetic induction, 
 all of which are capable of operating through distances 
 which are appreciable by our senses and measurable by our 
 instruments. On the other hand, it agrees thus far with 
 cohesion^ in virtue of which the molecules of bodies are held 
 together in masses, and with adhesion, which causes surfaces 
 of various kinds after being closely approximated to remain 
 united. 
 
 2. Though in a few cases solids may be made to com- 
 bine together by subjecting them to powerful pressure, or 
 by continued trituration, union occurs only at their surfaces 
 of contact, and cannot be completed unless one of the 
 bodies, at least, is in the liquid or in the gaseous state. That 
 is to say, the several parts of the acting masses must be free 
 
 H 
 
1 62 Chemical Philosophy. 
 
 to move, so that the mutual interpenetration required for 
 chemical combination may be perfect. 
 
 3. Moderate elevation of temperature generally favours 
 chemical combination, and is very frequently indispensable. 
 
 Thus it is a familiar fact that ordinary combustibles, 
 wood, coal, gas, &c., do not begin to burn unless the tem- 
 perature of some part of the mass is raised high enough. 
 Again, iron and sulphur or copper and sulphur have no 
 action on each other unless they are heated together very 
 strongly. Copper has no action on sulphuric acid unless 
 heated to about 130, and in many cases chemical actions 
 may be delayed or prevented altogether by keeping down 
 the temperature. 
 
 On the other hand, a large number of chemical com- 
 pounds can only be formed, and can only continue to exist, 
 at moderately low temperature. Thus many salts unite 
 with water, and ammonia unites with acids, to form definite 
 compounds which are destroyed by heat. Many facts of 
 this kind will be described in the chapter on * Dissociation,' 
 to which he reader is referred. 
 
 The production of polymeric compounds is a special 
 case of combination occurring under the influence of 
 moderate elevation of temperature. 
 
 Benzene, C 6 H 6 , for example, may be formed from 
 acetylene, C 2 H 2 (at a temperature below redness), and 
 terpilene, C 10 H, 6 , from isoprene, C 5 H 8 (at about 280). 
 In some of these cases depolymerisation, a reversal of the 
 process, occurs at a higher temperature. 
 
 4. The nature of the solvent or medium in which two 
 bodies are presented to each other often influences their 
 mutual action ; and the physical character of the products, 
 as to fusibility or volatility, frequently determines whether, 
 and in what manner, chemical action may occur. Thus, 
 when chloride of calcium and carbonate of ammonia are 
 dissolved in separate portions of water, and the solutions 
 then mixed together, a precipitate of carbonate of cal- 
 
Influence of the Solvent. 163 
 
 calcium is thrown down, and chloride of ammonium remains 
 in the mother liquid. But if these two salts, calcium 
 carbonate and ammonium chloride, are mixed together in 
 the dry state and heated, a decomposition occurs which is 
 the reverse of the last, and by which the original com- 
 pounds, chloride of calcium and carbonate of ammonia, are 
 re-generated. 
 
 Again, if acetic acid is poured into an aqueous solution 
 of potassic carbonate, effervescence ensues from the escape 
 of carbonic acid gas, and a solution of potassium acetate is 
 formed. But if potassium acetate is dissolved in strong 
 spirit of wine, a liquid in which carbonate of potassium 
 is insoluble, a stream of carbonic acid gas transmitted 
 through the solution is capable of decomposing the acetate, 
 throwing down a precipitate of potassium carbonate. 
 
 We may accept it as very generally, though not quite 
 universally, true, that when we mix together two soluble salts, 
 which by double decomposition are capable of giving rise to 
 an insoluble compound, that insoluble compound will be 
 precipitated until complete decomposition of one or both 
 the generating salts has taken place. If, for example, we take 
 the two salts, barium chloride and sodium sulphate, knowing 
 beforehand that barium sulphate is insoluble in water, we 
 may safely assert that a precipitate will be formed when the 
 aqueous solutions of these two salts are mixed together. 
 
 Similar observations hold good with regard to mixtures 
 of compounds which, amongst them, contain the elements 
 of a gas or body volatile at the temperature of the experi- 
 ment. In many cases this volatile body is formed. Con- 
 versely, if we submit to pressure or to a very low tempera- 
 ture a mixture of substances which, under ordinary atmo- 
 spheric conditions, evolves a gas, the chemical action is 
 retarded, or sometimes prevented altogether. 
 
 5. When several bodies capable of acting on one another 
 are mixed together, but the quantity of one of them largely 
 preponderates over the rest, some curious results are fre- 
 
 M 2 
 
164 Chemical Philosophy. 
 
 quently brought about. When, for example, a solution of 
 bismuth or antimony chloride in hydrochloric acid is mixed 
 with a little water, no change is perceptible to the eye, but 
 the addition of a larger quantity of water throws down a 
 white precipitate of the oxychloride, BiOCl or SbOCl. 
 These decompositions take place according to the following 
 equations : 
 
 BiCl 3 + OH 2 = BiOCl + 2 HC1 
 SbCl 3 + OH 2 = SbOCl + 2 HC1 
 
 That is to say, one molecule or 18 parts by weight of 
 water decompose one molecule or 314-5 parts of bismuth 
 chloride, or 226-5 P arts by weight of antimony chloride ; and 
 yet, if these proportions of the materials are brought into 
 contact, only a partial decomposition takes place, and the 
 reaction is completed only when a much larger quantity 
 of water is added. In such a change as this we must 
 remember that there are two antagonistic agents at work, 
 namely, the water tending to decompose the chloride ac- 
 cording to the equations given above, and the hydrochloric 
 acid which is generated by that decomposition tending 
 to reproduce the chloride. We may, therefore, consider 
 that there are four bodies in presence of one another, and 
 surrounded by water molecules. 
 
 Taking one case, 
 
 We have BiCl 3 , OH 2 , BiOCl, and HC1. 
 
 We may safely assume that when the number of water 
 molecules present is augmented, the number of molecules 
 of chloride decomposed by water in a given interval will 
 increase, whilst the number of oxychloride molecules de- 
 composed by the hydrochloric acid present will pari passu 
 diminish, until the whole of the chloride is destroyed and 
 precipitation is complete. 
 
 If now to the liquid holding the precipitate in suspension 
 we add hydrochloric acid in excess the action is reversed, 
 and the precipitate disappears again. 
 
Action of Mass. 165 
 
 In such experiments as these the decomposition seems to 
 proceed continuously as the quantity of the acting body is 
 increased, any alteration in the proportions, however small, 
 apparently producing a corresponding alteration in the extent 
 to which decomposition takes place. 
 
 Similar actions and reactions must be supposed to ensue 
 when salts in solution are mixed together, or when acids 
 are mixed with salts, with alcohols, or with ethers, and very 
 generally when several substances meet together, the final 
 condition of equilibrium being attained only after a lapse 
 of time which in some cases is long enough to be measur- 
 able, and the quantities of the several products varying ac- 
 cording to the relative masses of the active ingredients and 
 according to the relative intensity of their affinities. 
 
 6. The action of two substances upon each other is 
 often promoted by the presence of a third substance which 
 has the power of combining with one of the products of 
 change. In the ' basic ' processes for the production of steel, 
 pig-iron containing phosphorus is melted in a vessel lined 
 with calcined dolomite, and some common lime is generally 
 added, whilst air is either blown through the molten metal 
 or over the surface. Under these circumstances the oxidised 
 phosphorus unites with the lime, forming a phosphate, which 
 remains undecomposed by contact with the iron. Whereas 
 if the lining consists of an ' acid ' substance like silica, as 
 in the original form of the Bessemer process, phosphate of 
 iron is formed, which is immediately reduced by the heated 
 iron, and the phosphorus is thus returned into the metal, 
 and cannot be removed. 
 
 A somewhat similar action may be supposed to occur in 
 the production of sodium sulphate by Hargreaves' process. 
 Sulphur dioxide, oxygen, and water vapour react upon one 
 another very slowly to produce sulphuric acid, but when 
 these gases are brought into contact with common salt at 
 about 400 sodium sulphate and hydrochloric acid are freely 
 produced. 
 
1 66 Chemical Philosophy. 
 
 Another example of the action of a third substance is 
 afforded by the common methods for producing iodo-substi- 
 tution compounds. (See p. 222.) 
 
 In each of the foregoing cases the explanation of the 
 effect is tolerably obvious. There are, however, a great 
 many similar changes, the rationale of which has not yet 
 been made out. 
 
 Deacon's process for making chlorine viz. by passing a 
 mixture of air and hydrochloric acid over a heated copper 
 salt is one of these. So also is the action of a small 
 quantity of acid in promoting the combination of turpen- 
 tine with water to form terpin 
 
 CioH l6 + 2H 2 O = C 10 H.2oO 2 ; 
 the hydrolysis of such compounds as cane sugar 
 C|)H 3S O t | + H 2 O = C 6 H 12 O 6 + C & H 12 O 6 ; 
 
 Dextrose. Laevulose. 
 
 and the elimination of the nitrogen of cyanides and iso- 
 cyanides 
 
 HCN + H 2 = CH a Oa + NH ? . 
 
 Formic acid. Ammonia. 
 
 Another example is afforded by the application of alu- 
 minium chloride in promoting union of hydrocarbon radicles; 
 as, for, instance, when added to a mixture of benzene and 
 ethyl chloride 
 
 C 6 H 6 + C 2 H 5 C1 = HC1 + C 6 H 5 .C 2 H 6 
 
 Ethyl benzene. 
 
 In all these cases there is strong probability in favour 
 of the hypothesis that the action proceeds through a cycle, 
 an unstable compound being formed in small quantities at 
 a time and subsequently decomposed, the 'catalyst,' as the 
 active substance may be called, returning finally to the same 
 state of combination as at first. 
 
 Certain decompositions, which proceed slowly under the 
 action of heat alone, are also effected more quickly or at a 
 
Catalytic Actions. 167 
 
 lower temperature in the presence of a small quantity of a 
 second substance which apparently remains unchanged at 
 the end of the process. The most familiar example of this 
 is the decomposition of potassium chlorate by heat in the 
 presence of a small quantity of manganese dioxide, ferric 
 oxide, or cupric oxide, also the evolution of oxygen from 
 calcium hypochlorite by the action of a little oxide of cobalt. 
 
 In such cases as these, again, there is doubtless alternate 
 oxidation and deoxidation of the one substance by the other. 
 
 But perhaps the most remarkable examples of the action 
 of a minute quantity of a third substance upon the progress 
 of action between two others are found in the observations 
 which have been made within the last few years upon the 
 combustion of certain gases in oxygen. 
 
 It has been found by Professor Dixon that a mixture of 
 dry carbonic oxide and oxygen does not explode when an 
 electric spark is passed through it, but the addition of a 
 minute trace of water vapour makes the mixture explosive, 
 and the rapidity of the explosion increases with the quantity 
 of steam introduced. The presence of a minute quantity 
 of various gases which contain hydrogen, such as hydrogen 
 sulphide and chloride, ether vapour, or pentane, has a similar 
 effect ; but the addition of dry nitrous oxide or carbon di- 
 sulphide was not found to render the mixture inflammable. 
 
 Why steam and hydrogen compounds, which by inter- 
 action with oxygen produce steam, should be necessary for 
 the propagation of the flame through the gas has not been 
 fully explained. Probably at the temperature of the spark 
 the water molecule is less stable than the oxygen molecule, 
 and the decomposition expressed by equation I occurs. 
 
 I. CO + H 2 O = CO 2 + H 2 
 
 IL 
 
 The second phase may be an example of the effect of 
 two attractions acting concurrently, the hydrogen and car. 
 
1 68 Chemical Philosophy. 
 
 bonic oxide in equation II dividing a molecule of oxygen 
 between them, and the H 2 O then reacting with another 
 molecule of CO, and so alternately. 
 
 7. The peculiar activity of substances in the * nascent 
 state' has been already described (Chap. XV., p. 126), as 
 well as some other examples of concurrent chemical attrac- 
 tions. 
 
 CHAPTER XVIII. 
 
 DISSOCIATION. 
 
 PEROXIDE of nitrogen is a compound which exhibits very 
 remarkable phenomena. Below 10 it forms colourless 
 crystals which melt at higher temperatures, forming a yellow 
 liquid. This liquid boils at about 22, though, like water, 
 it evaporates at temperatures below this. The vapour is 
 yellow, but when gradually heated it first deepens in colour, 
 till at 100, and above, it becomes dark orange-red. If the 
 temperature is still further raised to 400 500 it suddenly 
 becomes colourless. On cooling these changes proceed in 
 reverse order, the dark-coloured gas being first formed, then 
 the pale yellow gas, ultimately condensing to a pale yellow 
 liquid. At the lowest temperature at which the density of 
 the vapour has been determined, namely, at 4-2, it was 
 found to be 2-588 (air = i), whilst the formula N 2 O 4 re- 
 quires the vapour density to be 3 -178. At about 150 the 
 density was found to be 1*58, which closely agrees with the 
 value i % 59 corresponding to the formula NO 2 . The loss of 
 colour at higher temperatures is doubtless due to the forma- 
 tion of nitric oxide, NO, and oxygen, both colourless gases, 
 though the vapour density in this condition has not been 
 ascertained. 
 
 A lump of common * sesquicarbonate ' of ammonia, or 
 
Dissociation or Thermolysis. 169 
 
 a solution of the same salt, exposed to atmospheric air at 
 common temperatures, loses ammonia, recognisable by its 
 pungency, and carbon dioxide, till ultimately no residue re- 
 mains. Or when a solution of ammonium oxalate, chloride, 
 or nitrate is boiled, a similar escape of ammonia may be 
 observed, and the liquid acquires a distinctly acid reaction. 
 
 (NH 3 ) 2 H a C 2 4 = NH 3 H 2 C 2 4 + NH 3 
 
 Ammonium Ammonium 
 
 oxalate. acid oxalate. 
 
 And if in either of these cases the ammonia thus evolved 
 is brought into contact at a lower temperature with the 
 solution from which it was produced, it will enter again 
 into combination, and the original compound will be re- 
 generated. 
 
 Again, when calcium carbonate is heated strongly in a 
 vessel from which the air has been more or less completely 
 removed by the air-pump, it suffers decomposition into lime, 
 which remains behind, and carbon dioxide gas, which fills 
 the vessel, CaCO 3 = CaO + CO 2 ; and this decomposition 
 proceeds until the evolved gas acquires a certain density or 
 tension, which increases as the temperature rises. If now 
 the whole is allowed to cool, the carbonic acid gas slowly 
 recombines with the lime, and a vacuum is once more 
 established. Calcium carbonate splits up in the same 
 manner when a current of steam, or other gas into which 
 the carbonic anhydride can diffuse, is passed over it whilst 
 under the action of heat. 
 
 Decompositions like that of nitric peroxide gas, of the 
 ammonium salts, or of calcium carbonate under the influ- 
 ence of heat, are examples of what is known as dissociation^ 
 or, as it is sometimes more precisely termed, thermolysis. 
 The word dissociation has acquired by careless use some 
 degree of ambiguity, but in these pages its application will 
 be restricted solely to those cases of decomposition in which 
 certain bodies are resolved at an elevated temperature into 
 simpler bodies, which are capable of reuniting and repro- 
 
170 Chemical Philosophy. 
 
 ducing the original compound when the temperature is 
 again allowed to fall. 1 
 
 We shall now consider several other cases of dissociation. 
 
 A crystal of sulphate of copper remains unaltered at 
 ordinary temperatures in moist air ; but if moderately heated, 
 it loses its water of crystallisation, and crumbles down to a 
 white powder. 
 
 The same kind of decomposition takes place in many 
 cases at the temperature of the air, and under ordinary con- 
 ditions. Salts, which thus readily part with their water 
 of crystallisation and fall away to powder, are said to be 
 efflorescent. Sulphate, carbonate, and phosphate of sodium 
 afford examples of this kind of dissociation, which, however, 
 presents nothing remarkable beyond the fact of occurring 
 at comparatively low temperatures. In this respect, how- 
 ever, these phenomena are surpassed by those exhibited by 
 salts, which, under ordinary circumstances, are deposited 
 from solution void of water of crystallisation. As already 
 mentioned (Chap. II., p. 15), such compounds have been 
 found in every instance to combine with water when crystal- 
 lised at temperatures below zero. The explanation of the 
 existence of anhydrous crystals is simply that dissociation of 
 the salt from the water occurs at or below the temperature 
 at which the crystals are deposited. And even those salts 
 which usually combine with water of crystallisation may be 
 obtained in a lower state of hydration, or altogether destitute 
 of water, by causing them to crystallise at more or less elevated 
 temperatures. Sodium sulphate furnishes a case in point. 
 The crystals of this salt formed at ordinary temperatures 
 contain ten molecules of water of crystallisation combined 
 with one of the salt (Na 2 SO 4 .ioOH 2 ) ; when formed at 
 18 they contain seven molecules of water (Na 2 SO 4 .7OH 2 ), 
 whilst a solution heated to 34 yields crystals which contain 
 
 1 The decomposition of ferric salts and other similar compounds, 
 when heated with water, ought not to be represented as cases of disso- 
 ciation, being in reality the effects of mass. (See Chap. XVII.) 
 
Abnormal Vapour Densities. 171 
 
 no water at all. Similar phenomena are beautifully exhibited 
 when the solutions of some coloured salts are heated. 
 Chloride of cobalt especially lends itself to this kind of re- 
 action. This salt forms red crystals consisting of CoCl 2 + 
 6OH 2 , and when dissolved in water it gives a red solution ; 
 but if the temperature is raised even very slightly the liquid 
 changes in colour, becoming successively purple and blue, 
 and these changes probably correspond with the formation 
 of increasing quantities of the compounds CoCl 2 +4OH 2 
 and CoCl 2 + 2OH 2 respectively. When the temperature 
 is again allowed to fall, the solution reassumes its original 
 red colour. 
 
 In the early attempts to apply the law of Avogadro to 
 the determination of molecular weights, many cases of 
 * abnormal vapour density ' presented themselves and gave 
 rise to much controversy. Thus sulphuric acid is entirely 
 vaporisable at a moderate temperature, but the relative 
 density of the vapour of this compound is only about one- 
 fourth, and not one-half, of the molecular weight as repre- 
 sented by the formula H 2 SO 4 , which accords with the 
 bibasic character of sulphuric acid. In other words, vapour 
 of sulphuric acid instead of being, in accordance with the 
 general rule, forty-nine times as heavy as an equal volume 
 
 of hydrogen taken at the same temperature, is only ^=24^ 
 
 times as heavy or thereabouts. 
 
 Ammonium chloride, phosphorus pentachloride, and 
 calomel afford instances of the same kind. These anomalies 
 are explained by the hypothesis that when these compounds 
 are volatilised they are at the same time resolved by disso- 
 ciation into two distinct products ; sulphuric acid into sul- 
 phuric anhydride and water, ammonium chloride into 
 ammonia and hydrochloric acid, phosphorus pentachloride 
 into phosphorus trichloride and chlorine, calomel into mer- 
 curic chloride and mercury. And in these particular cases 
 
172 Chemical Philosophy. 
 
 direct experimental evidence can be adduced that these 
 products exist in the several vapours. 
 
 In the case of sulphuric acid and of ammonium chloride 
 advantage has been taken of the difference in the diffusi- 
 bility of the products into which these bodies are supposed 
 to dissociate. Thus the vapour of sulphuric anhydride is 
 much heavier and consequently less diffusible than vapour 
 of water, so that when sulphuric acid is heated for several 
 hours in a vessel with a capillary orifice, the water vapour 
 escapes more rapidly than the sulphuric anhydride, and the 
 latter gradually accumulates in the residue. Pentachloride 
 of phosphorus has also been resolved into free chlorine 
 and phosphorus trichloride, which may be to some extent 
 separated by diffusion. 
 
 In the case of sublimed calomel we can appeal to the 
 fact, well known to manufacturers, that it invariably contains 
 small quantities of corrosive sublimate, and of metallic 
 mercury, which have escaped recombination during the 
 cooling down of the vapour. 
 
 Direct proof that dissociation has occurred is, however, 
 not possible in all cases, though there is usually some col- 
 lateral evidence available in support of the natural inference 
 from the vapour density. An interesting case is presented 
 by the halogens, especially iodine. The vapour density of 
 this element at temperatures up to about 700 is 8-8 (air = i) ; 
 above 700 the density diminishes rapidly, being reduced at 
 1500 (the pressure being -3 atmosphere) to nearly half 
 this, or 4-6. The conclusion seems to be inevitable that at 
 the lower temperature molecules of iodine consist of two 
 atoms, 1 2 , whilst at the higher they dissociate into separate 
 atoms. 
 
 The phenomena of dissociation appear to result from 
 two opposite and reciprocal actions the one of decompo- 
 sition, the other of recombination proceeding simultaneously 
 At low temperatures, when the composition of hydric sul- 
 phate, for example, is represented by the formula H 2 SO 4 or 
 
Dissociation Explained. 173 
 
 H 2 O,SO 3 , the number of molecules decomposed, into H 2 O 
 and SO 3 , is exactly counterbalanced by the number of 
 molecules which are reconstituted in the same period by. 
 the reunion of these two substances. As the temperature 
 rises and the agitation of the molecules in the mass be- 
 comes more vigorous, the number of molecules which un- 
 dergo decomposition progressively increases, whilst the 
 recombination part passu continually decreases, till at length 
 a point is reached in which the SO 3 molecules and the OH 2 
 molecules become indifferent to each other, and recombi- 
 nation no longer takes place. When the temperature is 
 allowed to fall these processes are reversed, and for every 
 degree of temperature a certain definite relation subsists 
 between the decomposition and recomposition, so that a 
 kind of equilibrium is maintained. The compensating 
 process being at high temperatures annulled, decomposition 
 is complete the vapour consists throughout of a uniform 
 mixture of two different kinds of molecules, and consequently, 
 by the law of Avogadro, it occupies twice the volume it 
 would otherwise fill if dissociation did not take place. 
 
 The rate and extent of dissociation in a given case there- 
 fore, independently of the chemical character of the substance 
 concerned, depends upon at least three conditions ; namely, 
 (i) temperature, (2) pressure, and (3) the proportion of the 
 products of dissociation to the undecomposed substance. 
 If one such substance is introduced at starting, the dissocia- 
 tion may be much impeded or even prevented altogether. 
 Thus if phosphorus pentachloride is mixed with a sufficient 
 quantity of the trichloride, PC1 3 , it gives a normal vapour 
 density corresponding to the formula PC1 5 . 
 
 It is obvious from what has gone before that a consider- 
 able number of compounds can only exist in the liquid or 
 solid state, and never, so far as known, in the state of gas. 
 A parallel has been traced between the phenomena of dis- 
 sociation and evaporation, or, in other words, between the 
 separation of molecules and atoms of different kinds and the 
 
174 Chemical Philosophy. 
 
 separation of molecules of the same kind in passing under 
 the influence of heat or diminished pressure from the solid 
 or liquid to the gaseous state. Thus the vaporisation of 
 water from a crystallised salt seems to be a phenomenon 
 of the same order as the escape of water vapour from ice. 
 Further, the progress of the change from liquid acetic acid 
 (see next page) to the same in a perfectly gaseous state at 
 250 is gradual and continuous throughout, as also in such a 
 decomposition as that of calcium carbonate under the in- 
 fluence of heat. In all such cases the rate of vaporisation 
 is governed by temperature and pressure, and is attended 
 by the absorption of the energy of heat, and reversal of 
 conditions leads to a restoration of the original state of the 
 substance. 
 
 One remarkable instance of the formation of a chemical 
 compound by compression of a mixture of two gases, and its 
 destruction by diminishing the pressure whilst the tempera- 
 ture remains unaltered, is afforded by phosphonium chloride. 
 When equal volumes of PH 3 and HC1 are mixed at 14 no 
 visible combination occurs, but on applying a pressure of 
 twenty atmospheres white crystals are formed of PH 4 C1, 
 which disappear into gas when the pressure is reduced. 
 
 Among the cases of dissociation which are commonly 
 recognised it is advisable to distinguish those in which a 
 double decomposition must be supposed to occur between 
 two molecules from those in which the products are derived 
 from a single molecule. Thus in such examples as the 
 following there is evidently an exchange of constituents be- 
 tween two molecules of the same kind : 
 
 HI + HI = H 2 + I 2 
 
 H 2 O + H 2 O = 2H 2 + O 2 
 
 CO 2 + CO 2 = 2CO + O 2 
 
 SC1 2 +SC1 2 = S 2 C1 2 + C1 2 
 
 The following list includes some of the most important 
 cases of simple dissociation without double decomposition. 
 
Examples of Dissociation. 
 
 175 
 
 Molecule of substance before 
 dissociation 
 
 Products of complete 
 dissociation at t 
 
 f approxi- 
 mately. Pres- 
 sure i atmos. 
 
 I 2 vapour below 700 .- 
 
 I+l .... 
 
 I500 
 
 Br 2 vapour .... 
 
 Br + Br 
 
 I500 
 
 Cl 2 gas .... 
 
 C1 + C1 
 
 above 1 600 
 
 S 6 vapour at 482 
 
 S 2 +S 2 +S 2 . 
 
 940 
 
 N 2 O 4 vapour below o 
 
 N0 2 + N0 2 . 
 
 160 
 
 [N 2 S ] liquid . 
 
 NO 2 + NO vapour 
 
 all temps. 
 
 PCl 5 b.p. i6o-i65 . 
 
 PC1 3 + C1 2 . 
 
 300 
 
 SbCl 5 .... 
 
 SbCl 3 + Cl 2 
 
 
 [Hg. 2 Cl 2 ] solid . 
 
 HgCl 2 + Hg vapour 
 
 950 
 
 [Al 2 Cy, z solid . 
 
 A1C1 3 + A1C1 3 vapour . 
 
 800 
 
 [IC1J,, solid 
 
 IC1+C1 2 vapour . 
 
 all temps. 
 
 H 2 F 2 vapour at 20 
 
 HF+HF 
 
 
 [NH 4 C1]* solid . 
 
 NH,+ HC1 . 
 
 350 to 940 
 
 [NH 4 Br] w solid . 
 
 NH 3 +HBr . 
 
 4 4 oto772 
 
 [NH 4 I] solid . 
 
 NH.+ HI . 
 
 440 to 772 
 
 [PH 4 I] W solid . 
 
 PH, + HI 
 
 
 [(C 2 H 5 ) 4 NI] solid 
 
 (C 2 H 5 ) 3 N + QH 5 I vapour 
 
 all temps. 
 
 (tetrethylammonium iodide) 
 
 < (triethylamine + ethyl- 
 1 iodide) 
 
 
 [NH 4 HS> solid. 
 
 NH 3 + H^S . 
 
 57 
 
 [HoS0 4 ] liquid . 
 
 H 2 + S0 3 . 
 
 416 
 
 [C 5 H U C1] liquid . . 
 
 C 5 H 10 +HC1 
 
 300 
 
 (isoamyl chloride) 
 
 (amylene) 
 
 
 [SCI,]* liquid at -22 
 
 SC1 2 +C1 2 . . . 
 
 8 
 
 [C 2 H 3 C1 3 O,] M solid, boils 95 
 
 C 2 HC1 3 O + H,O . 
 
 100 
 
 (chloral hydrate) 
 
 (chloral) 
 
 
 [C 2 H 2 (C,H 5 )C1 3 O 2 ] solid . 
 
 C 2 HC1 3 O + C,H 6 O 
 
 116 
 
 ( chloral -alcoholate ) 
 
 (chloral + alcohol) 
 
 a 
 
 [C 2 H 4 O 4 + CH 2 O 2 ] = 4 vols. . 
 
 CH 2 O 2 + CH 2 O 2 + CH 2 O 2 
 
 220 
 
 (formic acid vapour at b.p. \ 
 
 101) . i 
 
 (formic acid) 
 
 
 [C 4 H 8 O 4 + C 2 H 4 O 2 ] = 4 vols. 
 
 | C 2 H 4 2 +C 2 H 4 2 + i 
 C 2 H 4 O 2 
 
 250 
 
 (acetic acid vapour at b. p. 1 1 9) 
 
 (acetic acid) 
 
 
 Solid compounds of salts > 
 
 
 
 with ammonia, e.g. [AgCl. 
 
 Salt + wNH 3 
 
 
 Salts united with water of, 
 crystallisation . 
 
 Anhydrous salt + H 2 O 
 
 
ij6 Chemical Philosophy. 
 
 CHAPTER XIX. 
 
 THERMAL AND ELECTRICAL EFFECTS OF CHEMICAL 
 ACTION. 
 
 EVERY chemical change, whether it occurs in one substance 
 alone or as a result of the contact of two or more substances, 
 is attended by a redistribution of the total energy of the 
 system. This is most commonly manifested in the form of 
 absorption or evolution of heat. There is, however, nothing 
 in this to distinguish chemical change from physical or 
 mechanical changes, such as fusion, vaporisation, or altera- 
 tion of density. 
 
 Suppose a mass of ice at o to be so gradually heated 
 that it melts, but the resulting water remains at the same 
 temperature. The amount of energy in the form of heat 
 thus supplied would raise the temperature of the same mass 
 of water from o to 80. If the mass of ice is a gram 
 we say that 80 thermal units, or 80 units of heat, or 80 
 calories, are required for its fusion. 1 In similar terms 100 
 units are required to raise the temperature of a gram of water 
 from o to the boiling-point under a pressure of one atmo- 
 sphere, and 537 units more to change it into steam. At a 
 much higher, unknown, temperature the disruption of one 
 gram of water gas into a mixture of hydrogen and oxygen 
 would require at least 3,780 units, and probably much more. 
 All these changes are reversible, and when reversed the 
 same quantities of heat are successively given out. 
 
 The more obviously chemical changes which have been 
 studied in relation to the thermal changes which attend 
 them may be ranged under the following heads : 
 
 i. Combinations and decompositions at low tempera- 
 
 1 Sometimes the unit of mass referred to is a kilogram. The 
 corresponding thermal unit may be called a large calorie. 
 
Thermal Effects of Chemical Action. 177 
 
 tures. 2, Combinations at high temperatures (combus- 
 tion). 3. Dissolution in liquids. 4. Allotropic and isomeric 
 changes. 
 
 It will be convenient to consider these separately, and 
 in the present chapter to attend chiefly to the first division. 
 The others will be referred to in their proper places. (See 
 Chaps. III., XX., XXI.) 
 
 For the forms of calorimeters, the methods of experiment, 
 and the calculation of results, books on physics must be 
 consulted. 1 
 
 The estimation of the heat evolved during the neutralisa- 
 tion of an acid by a basic hydrate is perhaps one of the 
 least difficult of performance, and most intelligible among 
 thermo-chemical measurements. When solution of hydro- 
 chloric acid, for example, is mixed with solution of soda 
 sufficient to neutralise it heat is evolved, the amount given 
 out, however, varying to an appreciable extent according to 
 circumstances, whereof the most important are (i) the 
 proportion of water present, and (2) the temperature of the 
 solutions when mixed. 
 
 It is obvious, therefore, that the heat produced results 
 from a number of changes, part of which may be physical 
 or mechanical, and part chemical, though in these cases, at 
 any rate, there is no means of distinguishing them. Hence 
 the true thermal result of such a reaction as 
 
 HC1 + NaHO = NaCl + H 2 O 
 
 is unknown. After all, however, the practical object is to 
 compare together the action of substances of the same kind 
 under circumstances as nearly alike as possible. 
 
 Thomsen obtained the following values for the heat of 
 neutralisation of the common acids. This expression 
 means the number of calories evolved when one molecule 
 of a monobasic acid, or an equivalent quantity of polybasic 
 acid, is mixed with one molecule of sodium hydroxide, both 
 
 1 Experimental details and results will be found in Muir's Thermal 
 Chemistry. 
 
 N 
 
1 7 8 Chemical Philosophy. 
 
 being previously dissolved in 100 molecules of water and 
 at the temperature of 16 to 18 C. 
 
 Heats of Neutralisation of Acids by Soda. 
 
 Calories 
 
 HC1 .... 
 
 1^,740 
 
 HBr .... 
 HI .... 
 HF .... 
 
 HNO 3 .... 
 
 - 13,748 
 13,721 
 . 16,272 
 13,617 
 
 H 3 POo . 
 
 ie;,i6o 
 
 HP0 3 .... 
 iH 2 S0 4 .... 
 
 . 14,510 
 . 15,690 
 
 HCHOo .... 
 HC 2 H 3 O 2 
 HC 2 H 2 ClOo . 
 HCoHCl 2 O 2 . 
 HC 2 C1 3 2 . 
 
 n>343 
 13,45 
 . 13,400 
 14,280 
 . 14,830 
 . 13,920 
 . 13,840 
 . 12,400 
 
 The heat of neutralisation of the majority of acids 
 examined ranges between about 10,000 cal. and 15,000 cal. 
 It will be noticed that hydrofluoric acid stands at the top of 
 the list ; that hydrochloric, hydrobromic, and hydriodic acids 
 are nearly equal to one another. It is also interesting to 
 notice the effect on the thermal equivalent of introducing 
 chlorine in place of hydrogen into acetic acid. 
 
 When a molecule of an acid in aqueous solution reacts 
 upon sodium hydroxide, the thermal change is nearly, but not 
 exactly, proportional to the quantity of soda measured as i, 
 2, 3, or 4 molecules, according as the acid is mono-, di-, tri-, 
 or tetra-basic. In the neutralisation of dibasic acids, such as 
 the following, for example, it sometimes happens that the 
 evolution of heat is greater on addition of the first molecule 
 
Thermal Effects of Chemical Action. 179 
 
 of soda, sometimes on addition of the second. Thus on neu- 
 tralising one molecule of sulphuric acid by soda the first mole- 
 cule of NaHO added gives 14,7500. ; the second molecule 
 gives 16,6300. (mean 15,690 0.). With sulphurous acid, on 
 the other hand, the first stage towards neutralisation gives 
 15,870 0., the second 13,100 0. (mean 14,485 0.). What the 
 explanation of this difference may be is not known. It has 
 been supposed to support the idea that the constitution of 
 the two acids may be different, sulphurous acid and others 
 such as carbonic acid, which behave in a similar manner, 
 being represented as monobasic. It is conceivable, though 
 not very probable, that sulphurous acid may be 
 
 HO-S-H 
 
 A similar formula for carbonic acid cannot, however, be 
 admitted. 
 
 A substance in a chemically free or unsaturated condition 
 may be compared to a body which has been raised to a 
 height. In order to get it into such a position, energy in 
 some form must be expended, but the whole of that energy 
 is recoverable in the form of heat or mechanical effect by 
 the descent of the body to its former level. So, when a 
 chemical compound is resolved into its constituents, me- 
 chanical work, or rather its equivalent in the form of the 
 kinetic energy of heat or electricity, is transformed into the 
 potential energy of chemical separation ; but when the 
 elements come together again in their original order and 
 proportion the same amount of energy is again available, 
 chiefly in the form of heat. 
 
 Whenever much heat is evolved, it is tolerably safe to 
 conclude that the resulting compound is a very stable one. 
 Thus when i gram of hydrogen combines with 35 \ grams 
 of chlorine the result is not only the production of 36^ 
 grams of hydrochloric acid gas, but a certain amount of 
 kinetic energy which becomes manifest in the form of heat. 
 
 N 2 
 
I So Chemical Philosophy. 
 
 The amount of heat is represented by 22,000 units. In 
 order to reverse this effect, and bring about the decomposi- 
 tion of the compound gas into its constituents, and restore 
 them to their original physical condition, precisely the same 
 amount of energy must be supplied. 
 
 Further, all chemical changes which result in the produc- 
 tion of more stable from less stable forms of matter, whether 
 by way of combination or of so-called decomposition, are 
 attended by the evolution of heat ; and, as matter cannot 
 pass into conditions of instability without the aid of external 
 forces, all spontaneous chemical changes tend towards 
 the production of that body or system of bodies, in the 
 formation of which the greatest amount of heat is given out. 
 This has been called the ' principle of maximum work.' 
 Such, for example, are the action of acids upon bases in 
 solution, the displacement of the halogens by one another in 
 their compounds, the displacement of metals by one another 
 in their salts, &c. 
 
 From the following examples it will be seen that, with the 
 thermal values as data, we can predict what will happen 
 if one of the rnetals is placed in a solution of the chloride of 
 one of the others, which of the oxides is most easily 
 reducible by the action of carbon or other agent, why 
 chlorine decomposes bromides and iodides, and so on. 
 
 Heat evolved in the Formation of One Formula-weight in 
 Grams. 
 
 Chlorides in Watery Solution. 
 
 MgCl 2 . . 186,930 
 
 ZnCl 2 . . 112,840 
 
 CdCl 2 . . 9 6 > 2 5 
 
 HgCl 2 . . 59,86o 
 
 FeCl 2 . . 99,95 
 
 CuCl 2 . . 62,710 
 
 |AuCl 3 . 18,174 
 
 Hydrated Oxides, Solid. 
 
 MgO,H 2 O . 148,960 
 ZnO,H 2 O . 82,680 
 CdO,H 2 O . 65,680 
 
 FeO,H 2 O . 68,280 
 CuOH 2 . 37,520 
 
Thermal Effects of Chemical A ction. 1 8 1 
 
 Chlorides. 
 
 KC1 . 101,170 
 
 Nad . 96,510 
 
 . 93,820 
 
 HC1 . 39,320 
 
 Bromides. 
 
 KBr . 90,230 
 
 NaBr . 85,580 
 
 iCaBr 2 . 82,889 
 
 HBr . 28,380 
 
 Iodides. 
 
 KI . 75> 020 
 
 Nal . 70,300 
 
 . 67,670 
 
 HI . 13,170 
 
 In other cases, however, where the products of the 
 change remain in the field of action the change may never 
 be completed, although it might be expected from the 
 several thermo-chemical values taken separately. Thus, 
 when an acid is added to the aqueous solution of a salt of 
 another kind, it by no means follows that decomposition 
 will be complete, even though the heat of neutralisation of 
 the acid added is much larger than that of the acid formed 
 by its action in the solution. Thus sulphuric acid added 
 in equivalent quantity to a carbonate causes its complete 
 decomposition, but added to a solution of a chloride the 
 salt is completely changed into a sulphate only when the 
 hydrochloric acid is removed in proportion as it is formed. 
 (See also Chap. XVIL, pp. 162 to 165.) 
 
 In general, then, products of chemical combination 
 contain less energy than the materials from which they are 
 formed, the difference being disposed of in the form of heat. 
 But in some few cases the reverse is true ; that is, the 
 compound contains more energy than its constituents in 
 the state in which they are usually known. Thus the 
 formation of hydriodic acid, of nitric oxide, of ozone, and 
 probably of acetylene and carbon bisulphide, is attended 
 by absorption of heat, which is given out if these substances 
 are decomposed. 
 
 Comparing together the thermal effects of causing one 
 gram of hydrogen to combine with equivalent quantities of 
 chlorine, bromine, and iodine in presence of water, heat is 
 in each case evolved. 
 
 H + Cl + 400 H 2 O gives 39,320 calories 
 H + Br + 400 H 2 O 28,380 
 H + I + 400 H 2 O 13,170 
 
1 8 2 Chemical Philosophy. 
 
 A large part of the effect is, however, due to the solution 
 of the resulting hydracids in the water, and if the value of 
 this is subtracted we get a different result. 
 
 Heat of formation Heat of forma- 
 
 in presence of water Heat of solution tion of gas 
 
 HC1 . . 39,320 17,32 = 22,000 
 HBr . . 28,380 19,940 = 8,440 
 HI . . . 13,170 19,210 = 6,040 
 
 The negative sign means that in production of hydriodic 
 acid from hydrogen and iodine gases absorption of 6,040 
 units occurs for every unit of hydrogen which combines. 
 The decomposition of hydriodic acid gas is attended by 
 evolution of heat. We know how prone this compound is 
 to decomposition, and that even a solution of it, provided it 
 is strong, forms a powerful reducing agent, iodine being set 
 free. 
 
 In very dilute solutions, however, its reducing powers are 
 much less, as may be inferred from the thermal value of its 
 formation in presence of much water. 
 
 But in such cases as hydriodic acid, after all due allow- 
 ances have been made for the heat which becomes latent in 
 consequence of the change from solid to liquid, or gas, the 
 actual absorption of heat which is observed must be attri- 
 buted to the fact that energy is used up in the decomposition 
 of the elementary molecules which precedes, or is simulta- 
 neous with, the formation of the molecules of the com- 
 pound. Thus, when hydrogen combines with iodine, the 
 change is not simply 
 
 H + I = HI 
 
 but is, strictly speaking, a double decomposition ; thus 
 HI I 
 
 .HI HI 
 
 - 1 / i ; 
 
 Now, if the heat required to effect the separation of H 
 from H, and I from I, is greater than the amount of heat 
 generated when 2H combines with 2!, then the result will 
 
Electricity and Chemical A ction, 183 
 
 be negative ; no heat will be given out, but a certain amount 
 must be supplied from external sources. 
 
 From this consideration it is apparent that even in those 
 cases of chemical action in which an evolution of heat is 
 actually observed, this heat represents only the surplus energy 
 remaining over after the breaking up of the molecules of the 
 original substances employed in the experiment. 
 
 Not only is heat generated when chemical action occurs, 
 but in at least a great number of cases a definite amount of 
 electricity is developed. Imagine a plate of zinc plunged 
 into a solution of hydrochloric acid. The chlorine unites 
 with the zinc, and leaves the hydrogen to escape in bubbles 
 from the surface of the metal. If now a plate of platinum, 
 which is not acted upon by hydrochloric acid, is immersed 
 in the same liquid, and connected with the zinc by a wire, 
 it will be observed that the hydrogen is no longer disen- 
 gaged from the zinc plate, but collects upon the platinum 
 in bubbles which rise to the surface of the liquid. Such a 
 combination of two metallic plates immersed in a liquid and 
 connected together outside constitutes a voltaic cell. A 
 number of these cells joined together forms a battery. If 
 the wire is cut in two, that part which proceeds from the 
 zinc can be shown, by a gold-leaf electroscope, to be charged 
 with negative electricity, whilst the wire from the platinum 
 is positive. If these wires are then attached to platinum 
 plates dipping into a solution of some salt, such as iodide 
 of potassium, which is easily decomposable, the elements of 
 the salt are separated, the metal going to the negative elec- 
 trode or terminal connected with the zinc of the battery, 
 and the non-metallic element making its appearance at the 
 electrode belonging to the platinum plate 1 of the battery, 
 that is at the positive pole. This process of decomposition 
 
 1 The elements which make their appearance at the negative pole 
 are often referred to as electro-positive, whilst those which collect at 
 the positive pole are electro-negative. The metals are generally electro- 
 positive, the non-metals electro-negative. 
 
1 84 CJumical Philosophy. 
 
 by the voltaic current is called * electrolysis. 1 It deserves 
 to be noticed that it can only take place when the body 
 to be operated upon is in the liquid state. Now, just as 
 the amount of heat evolved by a given chemical combina- 
 tion is definite and constant under the same circumstances, 
 so the electrical effect produced by contact of the same 
 bodies under the same circumstances is constant. When 
 electrolysis occurs in the body through which the current is 
 passing, the quantities of its constituents which are liberated 
 are always proportional to their chemical equivalents, and, 
 provided secondary actions are guarded against, these 
 quantities are also chemically equivalent to the materials 
 consumed in each cell of the battery. If, for example, in 
 each cell 32*5 grams of zinc are dissolved, the current 
 passing simultaneously through solutions of iodide of 
 potassium, bromide of potassium, acidulated water, and 
 sulphate of copper would set free in the 
 
 ist cell, 127 grams of iodine, 39*1 grams of potassium. 1 
 2nd 80 bromine, 39-1 
 
 3rd 8 oxygen, i hydrogen 
 4th 8 oxygen, 3175 copper 
 
 The power of effecting electrolytic decomposition ex- 
 hibited by different combinations of metals and liquids 
 depends entirely upon the chemical characters of those 
 bodies. The more energetic the chemical action the more 
 powerful will be the electro-motive force of the combina- 
 tion. 
 
 As to the rationale of the decomposition effected by the 
 current opinions differ ; 2 certain phenomena of electrolysis 
 seem to be at present inexplicable. Thus liquid carbon 
 dioxide, cyanogen, carbon disulphide, hydrochloric acid, and 
 stannic chloride, when perfectly free from water, do not 
 
 1 Hydrogen and caustic potash would, of course, be the actual 
 products at the negative electrode. 
 
 2 See Chap. XXII. 
 
Electrolysis. 185 
 
 conduct and are not decomposed, and even water itself 
 when quite free from dissolved salts seems to be in the 
 same predicament. On the other hand, certain salts, such 
 as the chlorides and iodides of lithium, calcium, barium, 
 lead, and silver, when melted, not only conduct but undergo 
 electrolytic decomposition very readily. 
 
 It is not easy to see in what respect melted silver chloride 
 differs in chemical constitution from hydrogen chloride or 
 stannic chloride, yet the former yields, by the action of a 
 current from one or two Grove's cells, metallic silver and 
 chlorine respectively at the two electrodes ; whilst the latter 
 not only offer great resistance, but show no signs of decom- 
 position when the most powerful current yet obtained is 
 passed through them. But the case is altogether different 
 if water is added. Aqueous solutions of hydrochloric acid 
 or of tin chloride are decomposed with the utmost facility 
 by very feeble currents. When solutions are electrolysed 
 both the dissolved substance and the water seem to undergo 
 decomposition, the relative proportions in which the products 
 make their appearance depending on the strength of the 
 solution, and apparently also on the strength of the current. 
 Thus a concentrated solution of hydrochloric acid yields 
 only hydrogen and chlorine ; a weak solution yields 
 only hydrogen and oxygen. In this latter case it may be 
 that chlorine is really liberated, but has time or opportunity, 
 when there are many water molecules around, to attack 
 some of them, reproduce hydrochloric acid, and liberate the 
 oxygen. Possibly, however, since water alone and hydro- 
 chloric acid alone resist the current, the real electrolyte may 
 be some compound of the two existing in the solution. (See 
 also Chap. XXII., p. 205.) From solution of copper sulphate, 
 copper is alone deposited at the negative pole when the 
 current is weak, hydrogen also making its appearance if the 
 current used is strong. 
 
1 86 Chemical Philosophy. 
 
 CHAPTER XX. 
 
 COMBUSTION. 
 
 THE burning of wood, coal, charcoal, and other matters 
 commonly employed as fuel is a process with which every 
 one is familiar. Chemists know that the production of fire 
 in the usual way is attended not merely by the consumption 
 or alteration of the fuel, but by changes in the surrounding 
 atmosphere, and that the presence o.f a sufficient supply of 
 air is an indispensable condition in the operation. They 
 explain the phenomena by stating that the process of burn- 
 ing consists essentially in the combination of the elements 
 of the combustible body with the oxygen in the air, so 
 much heat being developed that more or less of the solid 
 combustible and of the products of combustion are raised 
 to such a temperature that they emit light. Notwithstand- 
 ing, then, that in ordinary fires the coals disappear and seem 
 to be destroyed, they do in reality only evaporate away in 
 the form of carbon dioxide and water ; and if these products 
 could be collected and weighed, their weight would be 
 found to be made up of the united weight of the carbon 
 and hydrogen of the coal, and the oxygen which is taken 
 from the air. 
 
 The phenomena of combustion may be observed equally 
 well when other materials are employed. Thus copper 
 burns in vapour of sulphur, hydrogen will burn in chlorine, 
 whilst phosphorus and several metals become ignited spon- 
 taneously when introduced in the proper condition into the 
 same gas. In every such case the resulting product con- 
 sists of a compound of the body which is burned with one 
 or other of the constituents of the gaseous atmosphere which 
 surrounds it. From this it is evident that the terms ' com- 
 bustible ' and ' supporter of combustion,' as generally em- 
 ployed, involve an error if they are taken to imply any 
 
Theory of Phlogiston. 187 
 
 difference of function ; for that which in one experiment 
 occupies the position of combustible may be made the 
 supporter of combustion or atmosphere in another. It is 
 easy to show, for example, that not only will a jet of hydro- 
 gen burn in oxygen gas, but that a jet of oxygen burns 
 equally well when surrounded by hydrogen. 
 
 Previously to the discovery of oxygen by Priestley, and 
 the establishment of the modern theory of combustion by 
 Lavoisier at the close of the last century, a remarkable 
 theory had been for upwards of fifty years adopted by 
 chemists. This was the celebrated theory of phlogiston, 1 
 proposed by Stahl. 2 This phlogiston was supposed to be 
 a substance of great tenuity, which, by combining with 
 incombustible bodies, rendered them combustible. When 
 such bodies are burnt, it was imagined that the escape of 
 the phlogiston in a peculiar condition of vibratory motion 
 gave rise to the phenomena of fire. At the time the idea 
 was originally introduced little was known of the part which 
 the air plays in all ordinary burning. When accumulated 
 facts proved conclusively that bodies by burning increase in 
 weight, some attempts were made to prop up the theory by 
 assuming that the presence of phlogiston gave bodies light- 
 ness instead of weight. The merit of the idea, however, 
 lay not so much in providing an explanation of certain 
 special cases of combustion, as in referring all cases of burn- 
 ing to a common cause, and in showing that the property of 
 combustibility is capable of being transferred from one body 
 to another. Oxides of the metals, for example, were re- 
 garded as ashes, or ' calces,' of the metal left after the escape 
 of their phlogiston, which could be restored to them by 
 contact with heated charcoal, a body which was supposed to 
 be specially rich in the hypothetical inflammatory principle. 
 
 Whilst we believe that the presence of no substance such 
 as phlogiston is necessary for the production of fire, and 
 that during the manifestation of the phenomena of com- 
 
 1 Q\oyiffT6s, anything set on fire. 2 Died at Berlin, 1734. 
 
1 88 Chemical Philosophy. 
 
 bustion no loss of material occurs, yet it has been very 
 justly pointed out that bodies, when they burn, do in truth 
 part with something, and that is the potential energy or 
 power of doing work which belongs to a state of chemical 
 isolation. (See Chap. XIX.) 
 
 In order that combustion may commence in air the 
 temperature of combustible bodies must in general be 
 raised. The temperatures required in different cases are 
 very diverse. Thus phosphorus, which liquefies at 44, can 
 scarcely be melted in the air without inflammation. Carbon 
 disulphide vapour mixed with air takes fire if a glass rod 
 heated to about 150 is brought into contact with it. 
 Sulphur begins to burn at about 250 far below its boiling- 
 point j whilst carbon and many hydrocarbons require a red 
 heat. 
 
 The temperature produced when the process of burning 
 is once established is in general higher than that which is 
 requisite for the commencement of combination. This 
 difference is illustrated by the action of platinum upon a 
 mixture of hydrogen or coal gas and air. If a warm slip of 
 clean platinum-foil or a coil of plantinum wire is held in a 
 current of such mixed gases, the temperature of the metal 
 rises rapidly, in consequence of combination taking place 
 between those portions of the gases which are in immediate 
 contact with it, combination extending to the surrounding 
 mass only when the temperature reaches a certain point, and 
 the platinum is nearly white-hot. Similar phenomena may 
 be observed in other cases when the heat evolved in the 
 early stages of the process is allowed to accumulate. The 
 spontaneous ignition of phosphorus, of finely pulverulent 
 iron or lead (pyrophori), and of heaps of oily rags, may be 
 referred to this cause. 
 
 The exact temperature of flame is difficult to determine, 
 and is liable to vary. The temperature of a hydrogen flame, 
 burning in air, has been estimated at about 2080 C., but 
 when the flame is fed with pure oxygen its temperature rises 
 
Heat of Combustion. 
 
 189 
 
 to upwards of 4000 C. This is easily explained by the fact 
 that in atmospheric air the oxygen is mixed with four times 
 its bulk of nitrogen, which contributes nothing to the 
 chemical action, and : which, being raised to the same tem- 
 perature as the other gases present, consumes a great deal 
 of heat. A temperature still higher is produced when a 
 mixture of hydrogen and oxygen in due proportions is fired 
 in a closed vessel, so that the heated gases are not allowed 
 to expand. This expansion against atmospheric pressure is 
 work the performance of which involves the consumption of 
 heat. The temperature produced by the explosion of oxygen 
 and hydrogen in a closed vessel has been estimated at about 
 
 5250 c 
 
 But although the temperatures producible by the same 
 combustible under various circumstances are different, the 
 actual amount of heat evolved in the combustion of the 
 same weight of a given substance is always the same. This 
 statement can of course be accepted only on condition of 
 uniformity in the circumstances attending the experiment. 
 Thus it will appear from the table given below that, as in 
 the case of carbon, the different allotropic modifications of 
 the same substance may give rise to appreciably different 
 amounts of heat. 
 
 I. Units of Heat developed by Combustion of Equal 
 Weights of Elements in Oxygen. 
 
 Substance burned 
 
 Product 
 
 Grams of water heated 
 i C. by burning i gram 
 of each substance 
 
 Hydrogen .... 
 Carbon 
 a. Diamond 
 b. Natural graphite 
 c. Wood charcoal 
 Sulphur (native) 
 Phosphorus (common) 
 Zinc 
 
 Water 
 
 Dioxide 
 
 
 
 Pentoxide 
 Oxide 
 
 34,034 
 
 7,770 
 7,797 
 8,080 
 2,22O 
 
 5,747 
 
 I.-2-2Q 
 
 Iron ..... 
 
 Peroxide 
 
 1,582 
 
190 Chemical Philosophy. 
 
 II. Units of Heat evolved by Combustion of Atomic Weights. 
 
 
 
 
 Grams of 
 
 Name of element 
 
 Weight in 
 
 Product 
 
 water 
 heated iC. 
 
 
 grams 
 
 
 by the com- 
 
 
 
 
 bustion 
 
 Hydrogen 
 
 I 
 
 Water (liquid) . 
 
 34,034 
 
 ,, . . . 
 
 I 
 
 Hydrochloric acid (gas 
 
 22,000 
 
 
 
 I 
 
 Hydrobromic acid (gas) 
 
 8,440 
 
 Carbon 
 
 
 
 
 a. Diamond . 
 
 12 
 
 Dioxide (gas) 
 
 93, 2 4O 
 
 b. Graphite . 
 
 12 
 
 a > 
 
 93', 560 
 
 c. Charcoal . 
 
 12 
 
 
 96,960 
 
 Sulphur (native) 
 
 3 2 
 
 Dioxide (gas) 
 
 71,042 
 
 Phosphorus (com mon ) 
 
 3 1 
 
 Pentoxide (solid) 
 Pentachloride (solid) 
 
 178,157 
 107,740 
 
 Tin . " . '. '. 
 
 118 
 
 Dioxide (solid) . 
 
 
 ' 
 
 118 
 
 Tetrachloride (liquid) 
 
 1 22,' 880 
 
 The quantity of heat absolutely evolved also depends 
 partly upon the physical condition of the products of com- 
 bustion. Thus the number 34,034, which expresses the heat 
 evolved in the combination of one part by weight of hydrogen 
 with eight parts of oxygen, represents not only the heat of 
 chemical action, but the heat (amounting at the tempera- 
 ture of the experiment to about 5,500 units) which is pro- 
 duced by the liquefaction of the resulting nine parts of 
 steam. 
 
 This relation of the amount of heat evolved to the 
 physical state of the resulting compounds is further indicated 
 by the results exhibited in the following table : 
 
 Substance burned 
 
 Weight 
 burned 
 
 Product 
 
 Units of heat evolved 
 
 Tin . 
 
 Sn =118 
 
 SnOo 
 
 135,360 
 
 Stannous Oxide 
 
 SnO =134 
 
 SnO 2 
 
 69,584 X 2 = 139,168 
 
 Copper 
 
 Cu = 63-5 
 
 Cu O 
 
 3 8 ,304 
 
 Cuprous Oxide 
 
 iCu 2 O= 71 < 
 
 CuO 
 
 18,304 x 2 = 36,608 
 
 Graphite 
 
 C =12 
 
 C0 2 
 
 93.560 
 
 Carbonic Oxide 
 
 CO = 28 
 
 C0 2 
 
 67,284 X 2 = 134,568 
 
Problems on Combustion. 191 
 
 Here we find that when solid tin or copper is converted 
 into its highest oxide, the amount of heat developed is, 
 practically speaking, twice as great as the amount of heat 
 developed in the conversion of the lower into the higher 
 oxide. In other words, the two successive stages of oxida- 
 tion, both of which result in the formation of solid products, 
 are marked by the evolution of equal quantities of heat. 
 
 The case of carbon is different. In the first stage of 
 oxidation the process involves the conversion of solid 
 carbon into gaseous carbonic oxide, whilst in the second 
 stage the carbonic oxide which is burnt, and the carbonic 
 anhydride which is formed, are both gaseous. There is no 
 change of state. Hence the quantity of heat which is de- 
 veloped in the latter operation is nearly two-thirds, instead 
 of only one-half, the total quantity evolved in the formation 
 of the same weight of carbonic anhydride from solid 
 carbon. 
 
 In order, therefore, to calculate the actual amount of 
 heat obtainable by burning a given combustible, it is neces- 
 sary to take these and other circumstances into consideration. 
 The following examples, which are unencumbered by small 
 corrections, and in which it is assumed that no heat is lost 
 by radiation or conduction, will serve to indicate the general 
 nature of such calculations. 
 
 The combustion of i part by weight of wood charcoal 
 evolves 8,080 units of heat. That is to say, i kilogram of 
 charcoal would heat 8,080 kilos, of water from o to i, or 
 i kilo, of water from o to 8080. 12 kilograms of charcoal 
 produce 44 kilos, of carbonic anhydride, or i kilo, produces 
 3*67 kilos. ; and if the heat produced by the combustion is 
 communicated to this quantity of carbonic anhydride, and 
 
 not to water, the temperature would be --, or 2202, if 
 
 3* 6 7 
 
 the specific heat of carbonic anhydride were the same as 
 that of water. But the specific heat of carbonic anhydride 
 is only -2164, when that of water is i. Hence the tern- 
 
192 Chemical Philosophy. 
 
 perature of the carbonic anhydride is 2202 x * , or 
 
 2164 
 
 10175, when the carbon is burnt in oxygen. 
 
 Now if the combustion is performed in atmospheric air, 
 which contains 77 per cent, of nitrogen, much heat is con- 
 sumed in raising the temperature of this nitrogen. 
 
 The 2*67 parts of oxygen required for the combustion of 
 one part of carbon are accompanied by 8*93 parts of nitrogen, 
 the specific heat of which is "2438. Therefore, when the 
 combustion of carbon takes place in air, the temperature of 
 the resulting mixture of gases cannot be higher than 
 
 8080 _ o r 
 
 (3-67 x -2164) + (8-93 x -2438) = 272C 
 
 In practice, the temperature is not so high as this, partly 
 because some heat is lost by radiation, some by conduction 
 through the solid unburnt charcoal, partly because an excess 
 of atmospheric air over and above that actually required 
 mingles with the products of combustion, and partly also 
 because, in all probability, the specific heats are not con- 
 stant, but increase as the temperature is higher. 
 
 CHAPTER XXI. 
 
 ISOMERISM. 
 
 SEVERAL bodies, though differing more or less in properties, 
 may have the same composition. In such cases they are 
 said to be isomeric. The differences observed among 
 isomeric bodies sometimes extend only to their physical 
 characteristics, sometimes to their chemical properties. 
 Several cases require, therefore, to be considered. 
 
 PHYSICAL ISOMERIDES. 
 
 Sulphur, when crystallised from carbon disulphide, yields 
 rhombic octahedra, the relative density of which is 2-07 ; 
 
Physical Isomerism. 193 
 
 whereas, if melted and allowed to cool, it crystallises in 
 oblique rhombic prisms, having the relative density 1*98. 
 The prismatic variety soon changes spontaneously into the 
 octahedral, which is the stable form, at the same time 
 evolving heat. Sulphur is said to be dimorphous, as it 
 crystallises in two forms, and these two modifications are 
 often spoken of as allotropic states of the element. 
 
 Many other examples might be cited of the same sub- 
 stance assuming different crystalline forms, the change of 
 structure being almost invariably attended by differences of 
 density and solubility. One form is generally less stable 
 than the other, and sooner or later, especially under the 
 influence of change of temperature, is converted into the 
 permanent variety. 
 
 When, as in these cases, two bodies chemically alike 
 exhibit slight differences of physical characters such as 
 solubility, crystalline form, relative density, or action upon 
 light they may be regarded as one and the same sub- 
 stance, though more or less disguised, and such bodies may 
 be distinguished as physically isomeric. 
 
 As examples of physical isomerism we have the two 
 varieties of native calcium carbonate, arragonite and Iceland 
 spar, and the curious instance of change of colour accom- 
 panying change of crystalline structure in the two modi- 
 fications of mercuric iodide. Common cane sugar, which 
 melts at 160, is converted at that temperature into an 
 amorphous modification (barley sugar), which slowly recovers 
 its crystalline structure if kept for some months. 
 
 The difference in all these cases probably arises not from 
 any difference in chemical composition or constitution, but 
 from the various modes in which the molecules of the bodies 
 are associated together ; such modification being connected 
 with some peculiarity in the circumstances attending their 
 formation. 
 
 That a change of molecular structure is accompanied by 
 a corresponding change of physical characters is proved by 
 
 o 
 
1 94 Chem ical Philosophy. 
 
 the well-known fact that when a piece of glass is strongly 
 compressed, either by mechanical means or by suddenly 
 cooling it from a high temperature, it acquires a power of 
 polarising light, which it loses when the pressure is removed. 
 
 POLYMERIDES. 
 
 Bodies containing the same elements united in the same 
 proportion, but having different molecular weights, belong 
 to this class of isomerides. The following are some ex- 
 amples : 
 
 Sulphur boils at 440, and is converted into an orange- 
 coloured vapour, the density of which, when taken at about 
 500, is three times as great as it should be theoretically; 
 whilst at 1000 it is only 32 times as great as that of 
 hydrogen at the same temperature and pressure, thus con- 
 forming with the ordinary rule. There seem, therefore, to 
 be two varieties of the gaseous sulphur molecule, one of 
 which is polymeric with the other. They may be repre- 
 sented by the formulae 
 
 S 2 and S 6 . 
 
 Ozone, the molecule of which has been shown by various 
 facts and arguments to have the formula O 3 , or |O 2 , may 
 also be regarded as polymeric with ordinary oxygen, O 2 . 
 
 The two varieties of nitric peroxide, NO 2 and N 2 O 4 , have 
 already been described (p. 168). The latter is polymeric 
 with the former. 
 
 Of carbon compounds exhibiting similar relations, the 
 hydrocarbons of the C n H 2n , or olefine series, afford a pro- 
 minent instance. The formulae of these bodies are all mul- 
 tiples of the first, methylene, CH 2 , which, however, is not 
 known in the free state. Being at once polymeric and 
 homologous with one another, they exhibit a regular gra- 
 dation in their boiling-points and densities, and form similar 
 chemical compounds. 
 
Polymerism and Metamerism. 
 
 195 
 
 Name 
 
 Formula 
 
 Boiling-point 
 
 Ethene or ethylene . . 
 
 C 2 H 4 
 
 Gas, 
 
 
 
 liquefiable only under 
 
 
 
 great pressure 
 
 Propene or propylene 
 
 C,H. 
 
 -17-8 
 
 Tetrene or butylene 
 
 C,H S 
 
 3 
 
 Pentene or amylene 
 
 C 5 H 10 
 
 35 
 
 Hexene or hexylene 
 
 C.H ia 
 
 6870 
 
 &c. 
 
 &c. 
 
 &c. 
 
 Further illustrations are supplied by the two chlorides of 
 cyanogen, CNC1 and C 3 N 3 C1 3 ; by cyanic acid, CNOH, and 
 cyanuric acid, CaNgC^H^ ; also by the modifications of 
 aldehyd, C 2 H 4 O, paraldehyd, (C 2 H 4 O) 3 or C 6 H* 12 O 3 , metal- 
 dehyd, (C 2 H 4 O) n , and many others. 
 
 METAMERIDES. 
 
 We have now to consider a kind of isomerism which 
 occurs very frequently among carbon compounds. The 
 nature of the phenomenon will be understood by comparing 
 together several such compounds as the following, all of 
 which are represented by the same empirical formula : 
 
 1. Propionic acid, C 3 H 6 O 2 , is a crystallisable acid, 
 which, after melting, boils at 140. It is monobasic, form- 
 ing one salt only with each of the metals sodium, potassium, 
 and silver. Its rational formula may, therefore, be written 
 thus : HC 3 H 5 O 2 . 
 
 2. Ethylic formate is a colourless aromatic liquid, which 
 boils at 56. When heated with caustic potash it is 
 resolved into ethylic alcohol and potassium formate. This 
 mode of decomposition is recorded when we write the 
 formula thus : C 2 H 5 .CHO 2 . 
 
 3. Methylic acetate is a colourless, volatile liquid, which 
 also boils at 56, but when decomposed by an alkali it 
 yields wood spirit (methylic alcohol) and an acetate. So 
 its formula must be CH 3 .C2H 3 O 2 . 
 
 o 2 
 
196 Chemical Philosophy. 
 
 These three compounds, which have the same composi- 
 tion and molecular weight, but differ in the nature of the 
 products they yield, when decomposed or acted upon by 
 chemical agents, are said to be metameric with one another. 
 
 Many cases similar to the last two may easily be found 
 amongst ethereal salts (compound ethers), ketones, and 
 other bodies, the molecules of which consist of two com- 
 pound radicles united together by oxygen or a bivalent 
 group. Pairs of such bodies may be called reciprocal 
 metamerides, because the excess of carbon and hydrogen 
 in one of the radicles is made up for by a corresponding 
 deficiency in the other. General formulae for such pairs of 
 isomerides among the compound ethers might be written 
 thus : 
 
 C n H 2n _ jOj Q and C P H 8P _ iO) Q 
 
 ^p-t^p + 1 j '-'n- tl 2n + 1 ) 
 
 In many cases an extraordinary resemblance in physical 
 properties may be observed in comparing together two 
 bodies of this kind. 
 
 Another instructive example of metameric relations is 
 presented by the two classes of alcoholic cyanides. 
 
 1. Cyanides. By distilling ammonium acetate with phos- 
 phoric anhydride (a substance which has an extraordinary 
 affinity for water) the ammonium salt is converted into a 
 volatile body, long known as acetonitril, and boiling at 77. 
 
 NH 4 C 2 H 3 2 - 2 OH 2 = NC 2 H 3 
 
 When acted upon by boiling alkali it yields up its nitro- 
 gen in the form of ammonia, and re-generates an acetate. 
 
 NC 2 H 3 + OKH + OH 2 = NH 3 + C 2 H 3 (OK)O 
 
 This reaction is explained by supposing that in acetonitril 
 or methyl cyanide the two atoms of carbon are in direct 
 union with each other, as represented by this diagram : 
 
 N = C - CH 3 . 
 
 2. Isocyanides or Carbamines. These compounds under- 
 
Asymmetric Carbon. 197 
 
 go a different transformation when acted upon by hydrating 
 agents. Notwithstanding that they have the same composi- 
 tion as the cyanides, they are scarcely affected by alkalies, 
 though when boiled with diluted acids they are readily con- 
 verted into formic acid and bases, in which the nitrogen is 
 associated with part of the carbon. For example, methyl 
 isocyanide treated in this manner yields methylamine and 
 formic acid. 
 
 NC 2 H 3 + 2H 2 = NJ** + CHO.OH 
 1^.11.3 
 
 Methyl isocyanide. Water. Methylamine. Formicacid. 
 
 In these compounds, then, the nitrogen probably forms 
 the link between the two atoms of carbon, in the manner 
 shown by the following graphic formula : 
 CEEN -CH 3 
 
 One very remarkable kind of metamerism is exhibited 
 by those numerous carbon compounds which possess the 
 power of rotating the plane of polarisation of a ray of 
 polarised light, such as the sugars and allied compounds, 
 the varieties of tartaric acid, of amylic alcohol and their 
 derivatives. In all such compounds, the constitution of 
 which is known, it seems to have been established, accord- 
 ing to MM. Van t'Hoff and Le Bel, that the molecule 
 contains at least one atom of carbon, having its four units 
 of valency occupied in four different ways. Thus malic 
 acid contains 
 
 CO 2 H-CH(OH)-CH 2 -CO 2 H. 
 
 In this formula we see one of the carbon atoms (printed in 
 heavy type) combined with CO.OH, with H, with OH, and 
 with CH 2 . This atom of carbon is said to be asymmetric. 
 
 In a great many of these optically active compounds 
 pairs of isomerides are known which, whilst exhibiting almost 
 complete agreement in chemical properties and in most of 
 their physical characters, possess the power of rotating the 
 polarised ray in opposite directions. Thus the dextro- and 
 
198 
 
 Chemical Philosophy. 
 
 laevo-rotating tartaric acids have the same constitution, and 
 no difference between them can be exhibited by the usual 
 formula 
 
 C0 2 H-(CH.OH)-(CH.OH)-C0 2 H, 
 
 which contains two asymmetric carbon atoms. According 
 to the hypothesis of Van t'Hoff the four radicles united to 
 each asymmetric atom are placed at the angles of a regular 
 tetrahedron, the atom of carbon occupying the centre. 
 Two arrangements of the attendant radicles are then 
 possible, the one being a reflection of the other, but not 
 superposible upon it, as shown in the following diagram. 
 
 When a compound, the formula of which contains one 
 or more asymmetric carbon atoms, is found to be inactive 
 to the polarised ray, it is usually resolvable into equivalent 
 quantities of two active isomerides, the one rotating to the 
 right, the other to the left. Thus racemic acid is made up 
 of dextro- and laevo- tartaric acids, which may be dis- 
 tinguished thus : 
 
 C(H.OH.CO 2 H)-C(H.OH.CO 2 H) 
 C(H.CO 2 H.OH)-C(H.CO 2 H.OH) 
 
 A fourth acid, called mesotartaric acid, which has long 
 been known, is optically inactive, but not resolvable like 
 racemic acid. It may be represented on the same hypo- 
 thesis thus : 
 
 C(H.OH.C0 2 H)-C(H.C0 2 H.OH), 
 
 where the effect of one half of the molecule is compensated 
 by the other half. 
 
Allotropy. 199 
 
 ALLOTROPY. 
 
 Many of the elements are known to exist in the form of 
 two or more modifications, which are very different in 
 physical properties and to some extent also in chemical 
 behaviour. One or two examples have been already re- 
 ferred to under the head of physical isomerism and of 
 polymerism. It is probable that a closer examination of 
 the different cases of allotropy known among the elements 
 would enable us to range them all in one or other of these 
 classes. At present, however, our knowledge will not allow 
 us to adopt with any degree of confidence a final decision 
 upon this point. 
 
 In the earlier part of this chapter prismatic sulphur is 
 described as a physical or mechanical modification of the 
 octahedral form of the element, and this is probably correct. 
 But these two are not the only varieties of which this body 
 is susceptible. By heating melted sulphur to a temperature 
 of 240 to 250 it becomes extremely viscid, and if cooled 
 suddenly whilst in this condition the viscid consistency is 
 retained, and the product is a tough elastic solid quite 
 different in aspect from ordinary sulphur. In this state it 
 is insoluble in carbon bisulphide, a liquid which takes up 
 octahedral sulphur very freely. 
 
 After keeping a few hours it becomes brittle and crystal- 
 line, and recovers its solubility in the usual solvents. The 
 same change may be brought about in a few minutes by 
 plunging it into hot water. 
 
 in these transformations, and in the insoluble character 
 of this plastic sulphur, we may trace a close resemblance 
 to the modifications to which certain carbon compounds, 
 such as aldehyd, cyanic acid, and other bodies, are subject, 
 and which are known to be the effect of polymeric changes. 
 It seems not unreasonable to consider that the production 
 of plastic sulphur may be brought about in the same way. 
 
 Phosphorus presents us with an example of a somewhat 
 
2OO Chemical Philosophy. 
 
 similar kind. This element in its ordinary state is at 
 common temperatures a solid of waxy consistence, which 
 becomes brittle at low temperatures. Its relative density is 
 1*82. It melts easily, dissolves in carbon disulphide, and 
 by sublimation or solution it may be obtained in brilliant 
 crystals in the form of regular octahedrons and dodecahe- 
 drons. When this body is heated to a temperature ap- 
 proaching 250 it is slowly transformed into a dull red 
 powder or mass, of relative density 2 '14, which is no longer 
 soluble in carbon disulphide. It shows itself also in many 
 ways less inclined to enter into chemical combination than 
 common phosphorus, being far less easily inflammable and 
 oxidisable, and unaffected by hot alkaline solutions. 
 
 The explanation of this appears, at least very probably, 
 to be that the molecules of the ordinary phosphorus combine 
 together into more complex groupings to form the allotropic 
 molecules, and so expend part of their chemical energy. 
 
 Carbon is another element which assumes several distinct 
 forms, the relations of which are of great interest. We may 
 divide these various modifications into two distinct groups, 
 the crystalline and amorphous. 
 
 Crystalline carbon is dimorphous. In one form it con- 
 stitutes the diamond, which crystallises in octahedral forms 
 of the regular system, and has a relative density on the aver- 
 age of 3*3. The other is graphite, or, as it is frequently 
 called, plumbago or black-lead, the crystalline form of 
 which, hexagonal plates, is quite incompatible with the form 
 of the diamond. The average density of graphite is 2*2. 
 If now we review the results which have been obtained by 
 burning the different forms of crystalline carbon, we find 
 that, allowing for slight experimental errors, the amount 
 of heat evolved by the combustion of equal weights of 
 diamond and graphite are practically the same. Twelve 
 grams of each substance burnt in excess of oxygen disen- 
 gage enough heat to raise the temperature of about 93,300 
 grams of water one degree, or, as it is usually expressed, 
 
Allotropy. 201 
 
 93,300 units of heat are evolved. The exact numbers in 
 each case are as follows : 
 
 Diamond .... 93,240 
 Natural graphite . . . 93,560 1 mean 
 Graphite from iron . . 93,140)93,350. 
 
 The smallness of the difference observed would lead 
 one to the belief that graphite and diamond possess the 
 same atomic structure, and that they owe their peculiarities 
 to different arrangement of their molecules that, in short, 
 they belong to the class of physical isomerides, were it not 
 for some remarkable facts in connection with their behaviour 
 under the influence of chemical reagents. A mixture of 
 nitric acid and potassium chlorate has no action on the 
 diamond, even in the state of the finest dust, but under the 
 influence of this powerful oxidising mixture graphite is con- 
 verted into a yellow crystalline substance, called by Sir B. 
 Brodie, who discovered it, graphic acid. This compound 
 contains C n H 4 O 5 (Brodie), and when heated it decomposes 
 violently, leaving a black graphitic residue, which still re- 
 tains oxygen and hydrogen. 
 
 Amorphous carbon may be obtained by a great variety 
 of processes, and in each case the product exhibits more or 
 less distinctly marked peculiarities. 
 
 But neither vegetable nor animal charcoal, lamp-black, 
 coke, nor gas carbon yields by the action of potassic 
 chlorate any substance of the nature of graphic acid, but 
 only black soluble substances of indefinite composition. 
 Twelve grams of wood charcoal give out 96,960 units of 
 heat when burnt so as to form carbonic anhydride, and 
 other kinds of charcoal, when deprived as completely as 
 possible of hydrogen and oxygen, give numbers closely 
 agreeing with this. Taking the average heat of combustion 
 of crystalline carbon as 93,300 units, it is obvious that there 
 is too great a difference here to be fairly accounted for by 
 
2O2 Chemical PhilosopJiy. 
 
 the hypothesis of experimental error, and consequently that 
 there is some essential difference in the constitution of 
 crystalline and amorphous carbon. The question whether 
 this difference is sufficient to indicate a polymeric relation 
 between these bodies remains to be answered. Silicon and 
 boron form allotropes, which are analogous to those of 
 carbon, and concerning which the same questions may be 
 propounded. 
 
 The study of a great number of cases has led to the 
 discovery that the formation of any two isomeric bodies 
 always involves the consumption of different amounts of 
 heat. Also, that when these bodies are burnt, or otherwise 
 similarly decomposed, the disruption of their molecules is 
 attended by the evolution of different amounts of heat. 
 
 This is nearly equivalent to saying that in order to 
 produce equal weights of two isomerides different amounts 
 of work must be expended in the two processes, and that 
 different amounts of energy are stored up in the products. 
 
 How is this energy disposed of? According to one 
 view, and adopting the molecular theory, we may reply that 
 the energy is employed in communicating to some atom or 
 atoms within the molecule a new kind of motion whereby 
 it acquires new chemical functions, and this change we 
 figure to our minds, and render intelligible by the hypothesis 
 that in the transformation of a body into its isomeride the 
 position of certain atoms contained within the molecule is 
 changed. We endeavour to represent this change of func- 
 tion by altering the arrangement of the symbols which go to 
 make up the formula of the body. Examples of this will be 
 found freely scattered through these pages ; but as an addi- 
 tional illustration we might refer to an interesting case of 
 isomeric change observed by Hofmann. 
 
 It was found that methyl-aniline, 
 
 /CH 3 
 (C 6 H 5 )'-N/ 
 
Chemical Affinity. 203 
 
 by protracted heating to a high temperature, is converted 
 into toluidine, 
 
 /H 
 <CH 3 )-(C 6 H 4 )"-N/ 
 
 In this metamorphosis an atom of hydrogen and an atom 
 of methyl, CH 3 , appear to exchange functions, and in order to 
 record this exchange the formulae are written in the above 
 or some similar manner, though it by no means follows that 
 we are to infer an exchange of place. 
 
 But in expending energy upon some part of a molecule 
 it is not very probable that the energy of the molecule as a 
 whole remains unaltered. Still, it is conceivable that in 
 some cases it may be so, and a careful comparison of the 
 chemical properties of a great many pairs of isomeric 
 bodies would be of the highest interest, as in this way we 
 might arrive at a solution of the question whether bodies 
 possessing a great store of energy are really more active in 
 their chemical behaviour than others in the production of 
 which a smaller amount of energy has been consumed. 
 
 CHAPTER XXII. 
 
 THEORIES REGARDING THE NATURE OF CHEMICAL AFFINITY. 
 
 THE belief that chemical compounds are formed by the 
 union, not of masses, but of atoms of matter, is generally 
 adopted by chemists. But why they unite, why a given 
 atom possesses the power of selecting among other atoms 
 presented to it, or why it habitually associates with one, 
 two, three, or more atoms of another kind (valency), are 
 questions which in the present state of knowledge cannot 
 be answered save by hypothesis. 
 
 Before entering upon the discussion of various theories 
 which have been at different times proposed to account for 
 
2O4 Chemical Philosophy. 
 
 the phenomena of chemical combination and decomposition, 
 it will perhaps be well to recapitulate (see also Chap. XVII.) 
 the chief characteristics which are attributed to chemical 
 affinity. 1 
 
 1. It acts only at inappreciable distances, smaller than 
 any which can at present be measured. Atoms chemically 
 united are probably not in actual contact with each other, 
 but are separated by a distance which is at least several 
 times their own diameters. If this is true, the attraction 
 which they exert upon each other must be believed to 
 involve some action in the medium which fills the space 
 between them, as in the case of other attractions at a dis- 
 tance, such as that due to electrical or magnetic induction. 
 
 2. It is selective. 
 
 3. Chemical change, whether combination or decom- 
 position, is attended by a concentration or dissipation of 
 energy, indicated usually by absorption or evolution of heat. 
 
 4. Chemical affinity differs from gravitation not only in 
 not operating at all distances, but in the fact that two or 
 more atoms, having combined together, lose more or less 
 completely the power of attracting other atoms. Gravitation 
 acts not only between a given body and one other, but be- 
 tween that body and all others, so far as we know, in the 
 universe. 
 
 5. Chemical compounds are broken up by heat; also (6), 
 according to common belief, by electricity. 
 
 Three chief hypotheses have been suggested. The first 
 assumes that chemical combination is the result of an * attrac- 
 tion ' of a peculiar kind which operates only at very short 
 distances. As already stated, this implies a condition of strain 
 in the ' ether ' or medium intervening, and until the relation 
 of the intensity of the attraction to the distance between 
 
 1 The word affinity seems to have arisen from a notion current 
 among the early chemists, that when two bodies are capable of combin- 
 ing, there must be some resemblance or affinity between them, or that 
 they contain some principle common to them both. 
 
Electro -chemical Theories. 205 
 
 any two atoms is known there is nothing to distinguish this 
 hypothesis from the idea of universal gravitation. 
 
 The electro-chemical theory of chemical combination has 
 attracted much greater attention. Immediately after the in- 
 vention of the voltaic pile it was discovered that the passage 
 of a current of electricity through water and various metallic 
 salts, either dissolved in water or melted by heat, decom- 
 poses them, their constituents appearing at the opposite 
 electrodes ; that hydrogen and the metals are always liberated 
 at the negative, whilst oxygen, chlorine, and acidulous bodies 
 generally are found at the positive pole. Later, Faraday 
 established two chief quantitative laws, from which we know 
 that the amount of decomposition in a given case is directly 
 proportional to the quantity of electricity flowing through 
 the circuit, and that if the same current passes through a 
 succession of decomposable substances or ' electrolytes/ the 
 amount of each element liberated is proportional to its 
 chemical equivalent (Chap. XIX., p. 184). To separate a uni- 
 valent atom, such as Cl, from a compound requires a certain 
 fundamental unit of electricity, the value of which has been 
 approximately determined. To separate a bivalent atom, such 
 as oxygen, requires twice as much, whilst a trivalent atom re- 
 quires three times as much, and so on, multiples of the unit 
 charge being always employed, but never fractions of it. 
 Hence it has been suggested that electricity, like matter, is 
 atomic. A hypothesis relating to the mechanism of elec- 
 trolysis was devised by Grotthus in 1805, and since that time 
 has been very generally accepted. In a liquid, such as a 
 solution of hydrochloric acid, each molecule of the elec- 
 trolyte is supposed to be made up of two parts one having 
 a charge of positive, the other of negative electricity, in 
 virtue of which these constituents are united together. 
 According to this hypothesis, then, chemical affinity is 
 nothing but the attraction of oppositely electrified particles. 
 When a current is sent through such a liquid, the first effect 
 is the establishment of a great number of chains of mole- 
 
206 
 
 Chemical Philosophy. 
 
 cules in which the positive elements all face in one direction, 
 the negative elements in the other. Then the metal plates 
 at different potentials which form the electrodes attract 
 the atoms of opposite electric name from the two ends of 
 the polar chains, and detach them from the state of chemical 
 combination in which they previously existed. 
 
 The following diagrams serve to illustrate the successive 
 conditions supposed to prevail in the liquid. In the first 
 the molecules of the electrolyte are moving about in the 
 liquid mass in every direction. Then these molecules 
 
 arrange themselves into chains in the order shown in II, 
 and in III the disruption of the end molecules occurs, 
 and is followed by a change of partners throughout the 
 length of the chain. Thus the disengagement of hydrogen 
 at the negative electrode, and of an equivalent quantity of 
 chlorine at the positive, without any apparent change in the 
 intervening mass of liquid is accounted for. There are, 
 however, many difficulties attending this hypothesis which 
 have not yet been explained away. Some of these have 
 already been referred to in Chap. XIX. Thus it has been 
 
Theory of Electrolysis. 207 
 
 shown that pure anhydrous liquid hydrochloric acid is prac- 
 tically an insulator or di-electric, and is not decomposed by 
 the most powerful current yet applied. Pure water also 
 resists the current so powerfully, that it is even now doubtful 
 whether water entirely free from other substances is capable 
 of electrolytic decomposition. Nevertheless, when hydro- 
 chloric acid and water are mixed together, a liquid is ob- 
 tained which is decomposable with the greatest ease. Again, 
 the hypothesis assumes that the constituents of an electrolyte 
 are united in consequence of charges of electricity resident 
 in the atoms. If these charges are supposed to be inseparable 
 from the atoms, the latter, when liberated from a compound, 
 should cling persistently to the surface of the electrode. 
 This, however, is not the case, for such substances as hydro- 
 gen, oxygen, or chlorine either dissolve in the liquid or 
 escape in bubbles. On the other hand, if the charges of 
 the several atoms are given up to the surface of the electrode 
 at which the atoms are set free, the coupling of the atoms 
 into chemically neutral molecules of hydrogen, H 2 , oxygen, 
 O 2 , or ozone, O 3 , &c., cannot be referred to charges of elec- 
 tricity. Moreover, since all atoms of any one element, say 
 hydrogen, liberated by electrolysis, must be charged equally 
 with the same kind of electricity (otherwise they would not 
 all pass to the same electrode), there is no reason to expect 
 that pairs of them would cling together, but rather that they 
 would repel each other, as similarly electrified masses com- 
 monly do. Another objection to such an electrical theory 
 of combination, as long ago pointed out by Dumas, arises out 
 of the fact that the halogens, strongly electro-negative as they 
 usually are, can replace hydrogen, atom for atom, in carbon 
 compounds, and the physical and chemical characters of the 
 substitution juducts closely resemble those of the substances 
 from which they are formed. (See, however, p. 220.) 
 
 The third hypothesis relating to the nature of chemical 
 changes is based upon the idea of atomic motion, and upon 
 this is founded the view of electrolysis which has been pro- 
 
2O8 Chemical Philosophy. 
 
 posed by Clausius. As already explained (in Chap. I. and 
 elsewhere), there are good reasons for believing that the 
 molecules of liquids and gases subsist unceasingly in a state 
 of motion. Their agitation is increased by heat, diminished 
 by cold. In the course of this dance in which they are 
 engaged, and the numerous encounters which must occur 
 amongst them, it is conceivable that some of the molecules 
 get broken up into atoms or atomic groups, which for a 
 while wander about until they encounter some other atom 
 or atomic group with which they can unite. If the original 
 body was homogeneous, the molecules which are thus re- 
 produced are of the same kind as the original molecules. 
 So long as this work of reproduction goes on at the same 
 rate as the destruction that is, so long as, in a given interval, 
 the number of molecules decomposed and the number of 
 molecules recomposed is the same no change occurs in the 
 properties of the body, because the average composition of 
 the mass remains the same. In a mass of hydrochloric 
 acid gas, for example, it is conceived that if it were possible 
 to submit it to such a scrutiny, the greater part of the mass 
 taken at any instant would be found to consist of molecules, 
 each made up of an atom of hydrogen and an atom of 
 chlorine ; but that with these there would be associated a 
 certain number of free atoms of hydrogen and chlorine 
 interspersed amongst them. In the next instant many of 
 these free atoms would be seen yoked together again, whilst 
 their places would be supplied by the disruption of fresh 
 molecules. 
 
 A similar condition is supposed to prevail in a liquid, so 
 that the separation of the atoms of an electrolyte is sup- 
 posed to be an effect of collisions among the molecules of 
 the compound, whilst the action of the current is merely to 
 direct the already separated atoms to the opposite elec- 
 trodes. Thus an atom of hydrogen being within a certain 
 distance of the positive electrode may be supposed to 
 receive a charge of positive electricity, the possession of 
 
Chemical Affinity and Adhesion. 209 
 
 which causes it to be attracted to the surface of the negative 
 electrode, where it arrives in company with other atoms 
 similarly charged. On reaching the electrode the atoms 
 give up their charges to it, and then pair off to form 
 molecules. 
 
 Conformable with this hypothesis we have the facts 
 already mentioned (pp. 184-5), tnat many pure substances 
 are not electrolytically decomposed until they are dissolved 
 in a liquid such as water ; and that whereas metallic 
 conduction is diminished by rise of temperature, electro- 
 lysis is usually promoted by raising the temperature of the 
 liquid. Both these conditions may be assumed to favour 
 dissociation. If electrolysis is merely a kind of electrical 
 convection in which previously dissociated atoms alone are 
 concerned, the electrical hypothesis of chemical affinity 
 falls to the ground. 
 
 Whilst for the more stable forms of chemical combi- 
 nation, such as give rise to the hosts of oxides, chlorides, 
 salts, and carbon compounds, the law of definite proportions 
 forms a safe criterion by which the operation of chemical 
 affinity may be recognised, many facts have been dis- 
 covered, especially of late years, which tend to support the 
 view that chemical combination is not always distinguish- 
 able from mere adhesion ; in fact, that these two forms of 
 union merely represent the extreme cases of a long series 
 of phenomena which pass by insensible degrees into one 
 another. 
 
 A short statement of some of these facts may be appro- 
 priately mentioned in this place. 
 
 Chemical affinity is elective. So is adhesion. Thus 
 glass is wetted by water but not by mercury. Again, if a 
 drop of a solution containing a mixture of cadmium and 
 mercuric chlorides is placed upon filter-paper, and allowed 
 to spread, it will be found that the cadmium extends further 
 into the paper than the mercury, the"Jatter remaining in 
 the centre of the blot whilst the former spreads to the edge. 
 
 p 
 
2io Chemical Philosophy. 
 
 Chemical combination is often supposed to be indicated 
 by rise of temperature. But when fine insoluble powders, 
 such as silica, are wetted with water, alcohol, or benzene, a 
 rise of temperature is observed, although no definite com- 
 pound appears to be formed. 
 
 It might be imagined that the relatively greater stability 
 of chemical union would be a sufficient guide, if such facts 
 as the copious absorption of hydrogen by palladium, and 
 the condensation of bromine vapour, ammonia, and other 
 gases by porous bodies, and especially by hard charcoal, 
 did not show that so-called adhesion may be equally 
 persistent. The film of air which covers the surface of 
 glass cannot at common temperatures be removed even by 
 the mercury pump, which produces a very perfect vacuum, 
 unless at the same time the glass is heated to near redness. 
 The tenacity of the combination in this case greatly exceeds 
 that of many recognised cases of chemical combination. 
 Relative stability evidently fails as a test when we consider 
 the properties of such a series of compounds as 
 
 SC1 4 , PH 4 Br, PC1 5 , NH 4 HS, NH 4 C1, Hg 2 Cl 2 , CaCO 3 , SiO 2 , 
 
 in which we have at one end compounds, the constituents 
 of which cannot remain united even at usual atmospheric 
 temperature and pressure, whilst at the other end there 
 is very great stability. 
 
 Again, chemical compounds may be decomposed not 
 only by the operations usually considered to be exclusively 
 chemical, but also by agents commonly regarded as non- 
 chemical or mechanical. Thus lead acetate and lead 
 nitrate, tartar emetic, or ammonio-sulphate of copper, may 
 be completely deprived of their respective metals if a solu- 
 tion of each salt is shaken up with pure carbon from bone 
 charcoal. 
 
 Many porous substances, such as paper or plaster, are 
 also capable of removing water of crystallisation from such 
 compounds as crystallised cobalt chloride. 
 
Chemical Affinity and Adhesion. 211 
 
 Beginning, then, with ordinary cohesion, in virtue of 
 which similar particles are attached to one another, and 
 adhesion which is the name given to the unknown cause 
 of the union of dissimilar particles and passing in review 
 the various phenomena of absorption, occlusion of gases by 
 solids, and solution, we arrive by imperceptible degrees at 
 the compounds which have been long distinguished by 
 many chemists as * molecular ' compounds, in which entire 
 molecules have been supposed to be united together by a 
 special kind of chemical affinity. This distinction is, however, 
 now rapidly disappearing, as it is acknowledged that there is 
 no character by which compounds of water or of ammonia 
 with salts, of iodine with iodides, or of salts with one another, 
 can be sharply distinguished from those in which atoms 
 are supposed to be linked together by ordinary chemical 
 affinity. 
 
 The more stable forms of combination seem to result 
 from a comparatively close approximation of the constituent 
 atoms or molecules, which is indicated by differences of 
 density. Thus the specific volume of * water of halhydration ' 
 (as it was called by Graham) in the sulphates of the 
 magnesium group is less than the ' water of crystallisation,' 
 each additional molecule of which occupies a gradually 
 increasing volume. 
 
 The following table shows the mean specific volumes of 
 MSO 4 nH 2 O, where M is Cu, Mg, Zn, Ni, Co, Mn, or Fe. 
 
 Value of 
 
 Mean Specific Volume 
 of MSO 4 nH 2 O 
 
 Molecular Volume 
 of th mol. of H 2 O 
 
 
 
 44-8 
 
 
 I 
 2 
 
 111 
 
 107 - 
 I3-3 
 
 3 
 
 83-3 
 
 I4'5 
 
 4 
 
 987 
 
 I5-4 
 
 5 
 
 112-9 
 
 I4-2 
 
 6 
 
 130-0 
 
 I7-I 
 
 7 
 
 146-1 
 
 16-1 
 
 P 2 
 
212 
 
 Chemical Philosophy. 
 
 These last figures approach the specific volume of ice, 
 
 which is M^LWt; = JL = IQ . 6 that of liquid water 
 
 Dens. -92 
 
 being 18. 
 
 When these salts are dissolved in water the water of 
 crystallisation acquires the same volume per molecule as 
 the rest of the water in the solution. 
 
 Again, when lead replaces silver, or potassium replaces 
 sodium in the nitrate, or when chlorine replaces bromine or 
 iodine in combination with another element, contraction 
 occurs. And in general contraction is observed when an 
 element of reputed strong affinity (as indicated by the 
 amount of heat evolved when it enters into combination) 
 takes the place of one of reputed smaller affinity. The 
 following examples will suffice : 
 
 Molecular Weight 
 
 Density 
 
 Mol. Wt. _ J^ Q J y j 
 
 Dens. 
 
 AgCl = 143-5 
 iPbd 2 = 139 
 
 5-5I7 
 578 
 
 26-0 
 24-0 
 
 Nal = 150 
 NaBr = 103 
 NaCl = 58-5 
 
 3'45 
 2-952 
 2-148 
 
 43'5 
 
 34'8 
 27-2 
 
 As to the relative strength or force of affinity between 
 two atoms in different cases, there is much apparent conflict 
 of evidence. Common experience seems to indicate that 
 the strength of the affinity between different atoms varies 
 according to their nature. Thus an iodide is decomposed 
 by bromine or by chlorine with liberation of iodine. Here, 
 however, the student must be again reminded that the 
 change is not correctly represented as 
 
 HI+ Br = HBr + I. 
 
 And though at first sight it seems to depend upon the 
 tendency for H to combine with Br in preference to I, to 
 
Relative Strength of Affinity. 2 1 3 
 
 this must certainly be added the affinity of I for I, the inter- 
 action probably proceeding in two stages, thus : 
 
 HI + BrBr = HBr + IBr 
 HI + IBr = HBr + II. 
 
 Little can be inferred even from the replacement of one 
 metal by another, as of mercury by copper, copper by iron, 
 &c., in solutions of their salts, for in all these cases the 
 process is complicated by electrical effects, or by a difference 
 of physical condition between the materials and the pro- 
 ducts of their interaction, also by the relative masses of the 
 acting ingredients. 
 
 It has been supposed that the amount of heat evolved 
 during the formation of series of compounds in which only 
 one element varies might serve as a measure of the relative 
 affinities of the several interchangeable elements, as, for 
 example, in the interchange of metals or of halogens for 
 one another (pp. 180-181). But though unquestionably 
 differences are manifest in such cases it does not follow 
 that the different amounts of energy degraded in the process 
 of combination indicate that the atoms combining come 
 together with different amounts of force. It has been con- 
 jectured that the loss of energy may be partly due to changes 
 in the configuration or motion of the atoms themselves. 
 That the amount of heat liberated in a chemical process is 
 no satisfactory measure of the amount of chemical affinity 
 at work is shown by the results of the researches of Julius 
 Thomsen and others on the relative ' avidity ' of acids and 
 bases for each other in solution. From these results it is 
 supposed that although acids on neutralisation by an alkali 
 evolve heat in different amounts, as shown in the table on 
 p. 178, the order in which they stand in respect to heat 
 evolved and affinity for the base is not the same. Thus 
 Thomsen concludes that when chemically equivalent quan- 
 tities of soda, hydrochloric acid (or nitric acid), and sul- 
 phuric acid in aqueous solution react upon one another the 
 
214 Chemical Philosophy. 
 
 base is not divided between the acids equally, but in the 
 ratio of about 2 : i. The activity of hydrochloric acid has 
 also been compared with that of sulphuric acid in causing 
 hydrolysis of milk sugar, and with a similar result. So that 
 although 2 HC1 is equivalent in neutralising power to H 2 SO 4 , 
 the dynamical equivalent of H 2 SO 4 is HC1. The rate at 
 which various other chemical reactions proceed, such as 
 the etherification of alcohols by acids and the transformation 
 of amides into ammonium salts, has also been determined, 
 and the results seem to point to the conclusion that each 
 substance acts partly in virtue of its mass that is, the 
 quantity present in the mixture and according to a specific 
 rate dependent upon its ' affinity.' 
 
 There is a further reason why the heat evolved or 
 absorbed in a chemical process is not a strict measure of the 
 energy of combination, for even in those cases in which 
 there is no alteration of physical state, as from solid to 
 liquid, there is always a change of volume. In the present 
 state of knowledge no distinction can be made between the 
 chemical and mechanical parts of each process. 
 
 On the whole, the hypothesis that chemical combination, 
 as well as adhesion and cohesion, is due to a state of 
 motion of the atoms or molecules engaged in a given 
 operation deserves most consideration. This view conforms 
 with all that is known concerning the mechanical properties 
 of gases ; it explains the phenomena of dissociation, and 
 without it the influence of ' mass ' upon chemical processes 
 would be unintelligible. The relatively great stability of the 
 molecules of the so-called ' elements ' (e.g. N 2 , O 2 , &c.) in 
 which the compound atoms are alike seems also explicable 
 on the hypothesis that this stability is due to a similarity or 
 harmony of their motions rather than to anything in the 
 nature of an electrical charge. 
 
Exercises on Section IV. 215 
 
 EXERCISES ON SECTION IV. 
 
 1. The density of the vapour of ammonium chloride is said to be 
 abnormal. Explain this statement, and describe the experimental 
 evidence upon which it is based. 
 
 2. Give some examples of the influence of * mass ' in determining 
 chemical reactions, and explain them. 
 
 3. Ammonium carbonate and calcium chloride are dissolved in 
 separate portions of water, and the solutions are then mixed. Express 
 the result by an equation. 
 
 Ammonium chloride and chalk, both dry, are mixed together and 
 heated. Express the result by an equation. 
 
 How do you explain such apparently inconsistent phenomena? 
 
 4. Ferric oxide is heated in hydrogen. Iron is heated in steam. 
 What effects are produced in each case, and how can you explain such 
 apparently contradictory results ? 
 
 5. Referring to the table of heat of formation of chlorides (pp. 180- 
 181), determine which of the metals, zinc, cadmium, mercury, copper, 
 gold, will decompose aqueous hydrochloric acid, and range them in 
 their apparent order of affinity for chlorine. 
 
 6. What substances and what relative quantities of each will be 
 formed by passing the same current through solutions in water of 
 potassium iodide, copper sulphate, hydrochloric acid (strong), and 
 hydrochloric acid (dilute)? 
 
 7. A current of electricity is passed simultaneously through solutions 
 of cupric and cuprous chloride. How much copper and how much 
 chlorine are liberated from the cuprous chloride for every molecule of 
 cupric chloride decomposed by the current ? In what relation do these 
 quantities stand to the quantity of zinc consumed in each cell of the 
 battery, secondary actions being neglected ? 
 
 8. Define in a few words the terms allotropy, metamerism, poly- 
 merism. 
 
 9. What weight of water would be heated from o to i by the 
 combustion of I gram of charcoal in oxygen ? 
 
 10. What weight of water would be heated from o to 15 by the 
 combustion of I gram of hydrogen in chlorine ? 
 
 11. Calculate the temperature of combustion of phosphorus burn- 
 ing in air. 
 
 12. Heat of combustion of hydrogen . 34O34 units. 
 Latent heat of steam at 100 . . . 537 ,, 
 Specific heat of steam .... '475 unit. 
 Specific heat of nitrogen . . . . '2438 
 
 Composition of air, N 77, 023 parts by weight. 
 
216 Chemical Philosophy. 
 
 With these data find the temperature of the hydrogen flame burning 
 in air. 
 
 13. What weight of water at 100 would be turned into steam by 
 the combustion of 1,000 kilograms of coal containing 
 
 Carbon 907 per cent. 
 
 Hydrogen 4-3 
 
 Oxygen and nitrogen . . . 4*0 ,, 
 Incombustible ash . . .1*0 ,, ? 
 
 Assume the heat of combustion of C and H as given in the table 
 on p. 189. Latent heat of steam at ioo = 537. 
 
 14. The heat of combustion of I part by weight of marsh-gas is 
 13,060. What weight of ice will be melted by the heat produced by 
 the combustion of a litre of the gas measured at normal temperature 
 and pressure ? 
 
 Latent heat of water at o = 79. 
 
 15. What volume (calculated at o and 760 mm.) of producer gas 
 consisting of 
 
 Carbonic oxide . . . . 25 per cent. 
 
 Hydrogen 5 
 
 Marsh-gas 4 
 
 Carbon dioxide . . . . 6 ,, 
 Nitrogen . . . . .60 ,, 
 
 must be burned in order to convert 2,000 kilograms of water at o into 
 steam at 100 ? 
 
 Heat of combustion of marsh-gas = 13,000. 
 
SECTION V. 
 
 CHAPTER XXIII. 
 
 CLASSIFICATION OF ELEMENTS. 
 
 THE following list includes all the known elements except 
 some twelve rare metals, the consideration of which, obscure 
 as their characters are, is unnecessary in a work like this. 
 The numbers in the table represent the relative densities 
 of those which are solid, usually in the densest form in 
 which they are known. 
 
 I. Non-metallic or Oxygenic * Elements. 
 
 Gaseous : 
 
 O, N, F, Cl (liquid 1-33). 
 
 Liquid : 
 Br 2-96 
 
 Solid : 
 
 S (octahedral) 2 "07 
 
 P (red) . 2-2 
 
 Si (graphitic) 2-49 
 
 B (adamantine) 2-68 
 
 C (adamantine) 3^5 
 
 Se (crystalline) 4*5 
 
 1 4'95 
 
 1 Oxygenic = acid -producing. 
 
218 Chemical Philosophy. 
 
 II. Metalloids 
 
 or Imperfect Metals. 
 
 Gaseou : 
 
 
 H 
 
 
 Solid : 
 
 
 Zr 
 
 4-15 
 
 V 
 
 S'S 
 
 As 
 
 5'9 
 
 Te 
 
 6-25 
 
 Sb 
 
 6-8 
 
 Sn 
 
 7'3 
 
 Mo . 
 
 8-6 
 
 Bi 
 
 . . 9-8 
 
 W 
 
 . 17-6 
 
 U 
 
 18-4 
 
 Ti, Nb, Ta (?) 
 
 III. Metals or 
 
 Basigenic } Elements. 
 
 Li 
 
 -578 
 
 K 
 
 . . '865 
 
 Na . 
 
 97 
 
 Rb . 
 
 1-52 
 
 Cs . 
 
 1-88 
 
 Be . 
 
 17 
 
 Mg . 
 
 17 
 
 Ca 
 
 1-8 
 
 Sr 
 
 2-5 
 
 Al 
 
 2-6 
 
 Ba 
 
 4-0 
 
 Ga . 
 
 5*9 
 
 Ge 
 
 5'5 
 
 Cr 
 
 6-8 
 
 Zn 
 
 7-0 
 
 In 
 
 7*4 
 
 Basigenic = base-producing. 
 
Classification of the Elements. 219 
 
 Fe 7-8 
 
 Mn . . . 8-0 
 Ni 8-6 
 
 Cd . 87 
 
 Cu 8-9 
 
 Co ... 8-9 
 Ag To-53 
 
 ' Ro . . . ii'o 
 Pb 11-36 
 
 Ru . . . 1 1 -4 
 Pd n-8 
 
 Tl n- 9 
 
 Hg (liquid at o) 13*596 
 
 Au . . 19-34 
 
 Os . . .21-4 
 
 Ir . . .21-15 
 Pt 21-5 
 
 This division into three groups is adopted here for pur- 
 poses of convenience, but the student must not infer that it 
 is absolutely necessary. Indeed, as he goes on he will find 
 that, as in all attempts to classify the things of nature, it is 
 impossible to define precisely a border-line separating a 
 given class of bodies from all others. 
 
 DIVISION I. NON-METALS. 
 
 These elements, as a class, are characterised by no 
 generality of physical properties. At common temperatures 
 and pressures four are gases fluorine, chlorine, oxygen, 
 nitrogen ; one, bromine, is a liquid ; the rest are brittle 
 solids. Of these, iodine, sulphur, selenion, and phosphorus 
 are fusible and vaporisable ; the remaining three are dis- 
 tinguished by infusibility (?), by non- volatility, by abnormal 
 specific heats (Chap. XIV., p. 115), and by affording, in the 
 cases of graphitic carbon and silicon, the only notable 
 examples among the non-metals of electric conductivity. 
 
220 Chemical Philosophy. 
 
 In their chemical characteristics, however, there is toler- 
 able uniformity. They all combine with hydrogen, forming 
 gaseous or very volatile hydrides ; they also combine with 
 metals, often in several proportions ; their oxides are either 
 neutral and indifferent bodies, like carbonic oxide, or, the 
 great majority, anhydrides, which by uniting with water form 
 acids. 
 
 NON-METALS. CLASS I I THE HALOGENS. 
 
 Fluorine . F = 19 
 
 Chlorine . Cl = 35'5) n , T 
 
 Bromine . Br == 80 !- - = 81-25 
 
 Iodine . . I =127 J 
 
 These elements are characterised by a remarkable family 
 resemblance. The last three especially are constantly asso- 
 ciated together in nature in the haloid * salts of potassium, 
 sodium, &c., and in the ores of mercury, silver, and other 
 heavy metals. They also agree very closely with one another 
 in their general physical characters and chemical deportment. 
 
 At ordinary temperatures chlorine is gaseous, bromine 
 liquid, iodine solid; but bromine and iodine are volatile, and 
 yield heavy, coloured vapours, which, when largely diluted 
 with air, have nearly the same odour as chlorine. Each 
 forms with hydrogen a strongly acid compound, which under 
 ordinary conditions is a colourless, fuming, very soluble gas, 
 consisting of equal volumes of hydrogen and the vapour of 
 the halogen, united without contraction. The chlorides, 
 bromides, and iodides of the alkali metals crystallise in the 
 same form, and the isomorphous replacement of the one 
 halogen by another is observed in a great many other cases. 
 
 The following table exhibits the formulae of all the known 
 oxides and acids of chlorine, bromine, and iodine, from 
 which it will be seen that although there are many gaps to 
 
 1 oAs = sea-salt = common salt. 
 
The Halogens. 221 
 
 be filled up, perhaps by future research, the correspondence, 
 so far as it goes, is almost complete : 
 
 Halogen Oxides and Corresponding Acids. 
 
 C1 2 C1 2 3 (?) C10 2 
 HC1 HC1O HC1O 2 HC1O 2 |HC1O 3 HC1O 4 
 
 HC10J 
 
 No oxide of bromine known. 
 HBr HBrO HBrO 3 
 
 I 2 5 
 
 HI HIO (?) HIO 3 (?) HIO 4 (?) 
 
 or H 2 I 2 6 or H 5 IO 6 
 
 The differences exhibited by chlorine, bromine, and 
 iodine are strictly gradational ; chlorine, with the smallest 
 atomic weight, being most active, bromine next, and iodine 
 the least energetic of the three. These differences are 
 manifested by their relative activity towards the metals and 
 hydrogen, chlorine displacing bromine, and bromine dis- 
 placing iodine from such combinations. 
 
 Indications of the same differences are afforded by the 
 superior activity of chlorine as a bleaching agent, and by 
 the energy with which it replaces hydrogen in carbon com- 
 pounds. As in several other cases of nearly allied elements, 
 to be referred to hereafter, the chemical activity diminishes 
 in proportion to the increase of the atomic weight, and rise 
 of boiling-point and density. 
 
 The replacement of one or more atoms of hydrogen in a 
 hydrocarbon by an equivalent quantity of one of the halo- 
 gens produces a neutral substitution compound ; but if a 
 similar replacement is effected in the molecule of a body 
 which contains oxygen, the product not unfrequently pre- 
 sents well-marked acid properties. This is the case, for 
 example, with some of the derivatives of phenol. 
 
 This oxygenic tendency of the halogens is also indicated 
 by the destruction of basic character in the amines or com- 
 
222 Chemical Philosophy. 
 
 pound ammonias by the substitution of chlorine, bromine, 
 or iodine for their hydrogen, as is well shown by the chlo- 
 rinated derivatives of aniline. 
 
 Aniline, C G H 7 N, a powerful base, 
 
 Chloraniline, C 6 H 6 C1N, less basic than aniline. 
 
 Dichloraniline, C 6 H 5 C1 2 N, feeble base. 
 
 Trichloraniline, C 6 H 4 C1 3 N, neutral. 
 
 Iodine. Iodine presents one or two peculiarities which 
 deserve special notice, as they serve to remove it to some 
 slight extent from immediate association with the kindred 
 elements, bromine and chlorine. In the first place, its 
 affinity for hydrogen is decidedly less energetic than that of 
 either of the other two elements. This is indicated first by 
 the fact that iodine does not usually bleach vegetable 
 colours ; secondly, that, acting alone, it is incapable of 
 producing substitution derivatives from carbon compounds. 1 
 Whenever substitution of chlorine, bromine, or iodine 
 occurs, the hydrogen which is necessarily eliminated goes to 
 form the corresponding hydracid. Now, in the case of iodo- 
 substitution compounds, it has been shown that they are all 
 decomposed by the action of hydriodic acid, with repro- 
 duction of the original body and free iodine. Hence iodo- 
 substitution compounds cannot be formed by the action of 
 iodine, unless precautions are taken to remove or to destroy 
 the hydriodic acid that may be produced. This is effected 
 in various ways, usually by the action of mercuric oxide or 
 iodic acid. The difficulty may also be got over in some 
 instances by substituting iodine monochloride for iodine. 
 Thus, orcinol acted upon by a solution of iodine chloride 
 gives tri-iodorcinol and hydrochloric acid 
 
 C 7 H 8 2 + 3 IC1 = C 7 H 5 I 3 2 + 3HC1. 
 
 Another distinguishing characteristic of iodine is the in- 
 tense colour exhibited by the vapour of the element itself, by 
 
 1 See also Chap. XIX., p. 181, Heat of Combination of Iodine 
 with Hydrogen. 
 
Iodine. 223 
 
 its solutions in certain liquids, notably in carbon disulphide, 
 by its compound with starch, and by many iodides, the 
 corresponding chlorides or bromides being either colourless 
 or very pale. 
 
 Again, chlorine and bromine are more soluble in water 
 than iodine, and are even capable of forming at low 
 temperatures crystalline hydrates, having the formulae 
 Cl 2 .ioH 2 O and Br 2 . ioH 2 O, no such compound being 
 formed by iodine. On the other hand, the solubility of 
 chlorine and bromine is not increased by the addition of a 
 chloride or bromide to the water in which they are to be 
 dissolved. The solubility of chlorine in aqueous solutions 
 of chlorides is usually less than its solubility in pure water. 
 Iodine, however, is freely soluble in iodide of potassium, 
 and, indeed, produces in this way a black liquid which 
 contains an unstable tri-iodide of potassium, KI 3 . This 
 compound forms dark blue deliquescent prisms. The cor- 
 responding ammonium salt, NH 4 I 3 , resembles it closely, 
 and analogous periodides are formed by the organic ammo- 
 nium bases, some of which form crystals of great beauty, 
 containing two, four, six, or even eight atoms of iodine in 
 addition to the elements of the normal iodide ; as, for 
 example, the following compounds derived from tetramethyl- 
 ammonium iodide : 
 
 N(CH 3 ) 4 I, N(CH 3 )J 3 , N(CH 3 ) 4 I 5> N(CH 3 ) 4 I 9 
 
 The iodates exhibit some anomalies for which there is 
 no parallel among the chlorates. Thus, in addition to the 
 normal potassic iodate, KIO 3 , there are two other well-crys- 
 tallised salts, containing an excess of anhydride, for which 
 it is difficult to find analogues, except, perhaps, among the 
 chromates. 
 
 Iodates Chromates 
 
 KI0 3 
 
 or K 2 I 2 O 6 K 2 CrO 4 
 
 K 2 I 2 O 6 .I 2 O 5 K 2 CrO 4 .CrO 3 
 
 K 2 I 2 O 6 .2loO 5 K 2 CrO 4 .2CrO 3 
 
224 Chemical Philosophy. 
 
 This tendency of iodine to accumulate in its compounds 
 is just one of those characters which belong especially to 
 multivalent elements, among which iodine seems, on the 
 whole, entitled to be placed. 
 
 Fluorine. This element has been at last isolated (1886) 
 by the electrolysis of a mixture of hydrogen and potassium 
 fluorides. It is described as a colourless gas which decom- 
 poses water, forming hydrofluoric acid and ozone. Silicon 
 and boron burn in it, forming fluorides. Fluorine is sepa- 
 rated from the other halogens, partly in consequence of the 
 vapour density of hydrofluoric acid, corresponding to the 
 formula H 2 F 2 at temperatures much above its boiling-point, 
 partly by its very extraordinary attraction for silicon, partly 
 by the non-existence of any oxide or oxyacid of fluorine. 
 In spite, also, of their general resemblance to the chlorides, 
 bromides, and iodides, individual fluorides differ in many 
 cases from the corresponding chlorides. Thus, fluoride of 
 calcium is insoluble, chloride of calcium very soluble and 
 deliquescent ; fluoride of silver is soluble in water, chloride 
 of silver totally insoluble ; fluoride of potassium soluble in 
 water, but, unlike the neutral stable chloride, it yields an 
 alkaline solution which probably contains caustic potash and 
 the double hydrogen and potassium fluoride. 
 
 2 KF + OH 2 = KOH + KF.HF. 
 
 The tendency to produce double salts of this kind has, 
 indeed, led to the idea that fluorine may be really a bivalent 
 element, like oxygen, with the atomic weight 38. The 
 formulae of the double fluorides would then be comparable 
 with those of oxygen compounds. 
 
 Fluorides Oxides 
 
 H 2 F analogous to H 2 O 
 
 KHF KHO 
 
 KBF 2 KB0 2 
 
 K 2 SiF 3 K 2 Si0 3 
 
 K 2 SnF 3 K 2 SnO 3 
 
 &c. &c. 
 
Valency of the Halogens. 225 
 
 These double fluorides, however, are not more numerous 
 or prominent than are the double chlorides, bromides, and 
 iodides, and it seems not unreasonable to explain their 
 existence by a similar hypothesis. The atom F(= 19) may 
 be occasionally trivalent. If so, the constitution of fluoride 
 of potassium and hydrogen may be represented as 
 
 KF = FH, 
 
 and that of the other fluorides, single and double, in a 
 similar manner. 
 
 Valency of the Halogens. An atom of a halogen never 
 replaces, in a direct manner, more than one atom of hydro- 
 gen. In the haloid salts of these elements, 35^5 parts of 
 chlorine and equivalent quantities of bromine, iodine, and 
 fluorine are almost always combined with the metallic 
 representative of one part by weight of hydrogen. The 
 halogens are therefore generally univalent. Nevertheless, 
 many compounds are known, the existence of which can 
 scarcely be accounted for except upon the hypothesis of 
 their occasional trivalent function. 
 
 Thus, in addition to the normal iodides and iodo-substi- 
 tution compounds, 
 
 H'l, K'l, Hg"I 2 , C 2 H 3 r0 2 , 
 
 iodine forms the following compounds : 
 
 I'"C1 8J I'" (C 2 H 3 2 ) 3 , KAgI'" 2> 
 which may be written as follows : 
 
 Cl Cl\ /0-C 2 H 3 
 
 I Cl I- 0-C 2 H 3 O K-I=I-A g 
 
 I Cl or 
 
 S Cl Cl/ \O-C 2 H 3 
 
 In their oxygenated compounds, chlorine, bromine, and 
 iodine present also a very marked resemblance to nitrogen, 
 which is most usually (perhaps always) a triad. Thus we 
 
 Q 
 
226 Chemical Philosophy. 
 
 have hypochlorous and hyponitrous acids, both extremely 
 unstable bodies, known chiefly in the form of their salts ; 
 chlorous and nitrous acids, also very unstable ; chloric and 
 nitric acids, both liquid, easily decomposable, highly corro- 
 sive bodies, the salts of which are all soluble in water. The 
 nitrogen analogue of perchloric acid is at present unknown. 
 The formulae of these corresponding pairs of compounds 
 are as follows : 
 
 HC1O HC1O 2 HC1O 3 HC1O 4 
 
 HNO HNO 2 HNO 3 
 
 A further correspondence is observable in their oxides : 
 
 C1 2 O C1 2 O 3 (?) C1O 2 I 2 O 5 
 
 N 2 N 2 3 (?) N0 2 andN,O 4 N 2 O 6 
 
 Now, if we admit that nitrogen is trivalent in these com- 
 pounds, the presumption that chlorine and its congeners 
 are also trivalent is, at least, worthy of discussion. The 
 following graphic formulas express the constitution of chloric 
 and nitric acids upon this hypothesis : 
 
 H-0-C1/ | H-O-N/ | 
 
 Chloric acid. Nitric acid. 
 
 As contributing evidence in support of the same view, 
 we might also point to the large number of permanent and 
 definite double compounds which the chlorides, bromides, 
 and iodides are capable of forming. We have, for example, 
 the well-known chloro-platinates of the alkali metals 
 
 2NaCl.PtCl 4 
 2KCl.PtCl 4 , &c., 
 
 besides the innumerable compounds of the alkali-metal 
 chlorides with the chlorides of magnesium, iron, aluminium, 
 mercury, copper, and other metals. Although a few of 
 these compounds are decomposed by water, the majority of 
 them are stable enough, and give no signs of alteration by 
 
Oxygen. 227 
 
 such treatment. Their constitution is most reasonably 
 represented upon the assumption of the trivalence of the 
 halogen. The compound KI.HgI 2 , for example, may be 
 written in this manner : 
 
 I 
 
 Some chemists regard these compounds as formed by 
 the union of entire molecules of the constituent salts, com- 
 bined together by some kind of adhesion differing from 
 ordinary chemical affinity. Thus the potassio-mercuric 
 iodide represented above is considered to contain both 
 iodide of potassium and iodide of mercury, HgI 2 + KI. 
 This salt, however, is colourless, soluble in water, and totally 
 unlike the scarlet insoluble mercuric iodide, the elements of 
 which it contains, and, so far as physical characters are 
 concerned, there is nothing to distinguish such cases from 
 examples of ordinary atomic combination. 
 
 NON-METALS. CLASS 2. 
 
 = 79- 
 
 These elements are associated together principally by 
 reason of the correspondence in composition and general 
 properties among their compounds with hydrogen 
 
 OH 2 SH 2 SeH 2 TeH 2 . 
 
 For although, taking any two adjacent terms of the series, 
 many points of resemblance may be traced out, yet there is 
 a very wide interval between the colourless gas oxygen and 
 the silvery, metallic, crystalline solid tellurium. 
 
 Q 2 
 
 Oxygen . 
 Sulphur 
 Selenion . 
 Tellurium . 
 
 .O=i6 
 
 - S = 32) c 
 
 e p - n 1 k 
 
 5 + Te 
 
 . be 79 y 
 . Te = 126] 
 
 2 
 
228 . Chemical Philosophy. 
 
 Oxygen, standing apart from the rest in virtue of many 
 extraordinary qualities, deserves to be noticed first. This 
 element constitutes nearly one-half the total weight of the 
 earth's crust. Indeed, it seems to be the preponderating 
 element of the globe. 
 
 Leaving out of account the insignificant quantities of 
 metals and metallic sulphides embedded in the earth, it 
 may be said that the entire constituents of the earth's crust, 
 including the water of the ocean, consist of compounds 
 saturated with oxygen ; and even supposing all organic 
 bodies to be completely burnt up, the atmosphere would 
 still contain a considerable quantity of unemployed oxygen. 
 
 Oxygen is entirely unmatched among the rest of the 
 elements, both as regards the number and varied character 
 of its compounds, and the important part which it plays in 
 relation to combustion and life. 
 
 Water, hydrogen protoxide, also exhibits properties which 
 are in every way remarkable as specially fitting it for the 
 part it has to play in the economy of nature the very 
 type of a neutral body, yet capable, under special circum- 
 stances, of acting either as a feeble base or a feeble acid ; 
 an almost universal solvent of saline bodies ; exhibiting the 
 anomaly of attaining maximum density at a temperature 
 (4) above its solidifying point; standing alone amongst 
 liquids by reason of its great capacity for heat ; having a 
 vapour-density little more than half that of its gaseous sul- 
 phur analogue (SH 2 ), and yet liquefying and solidifying 
 readily. 
 
 The higher oxide of hydrogen, O 2 H 2 , with its bleaching 
 and oxidising powers, resembles in no slight measure the 
 element chlorine. 
 
 The name oxygen, 'acid producer,' from ovc and 
 yevi/aw, was given by Lavoisier under the mistaken impres- 
 sion that this element contained a principle common to all 
 acids. This we know was an error. Nevertheless, the 
 name was not ill-chosen, and for the following reasons : 
 
Oxygen and Oxides. 229 
 
 1. The majority of known acids contain oxygen. The 
 chief exceptions to this statement are as follows : 
 
 (a) The so-called hydracids, HF, HC1, HBr, HI, HCN, 
 and their compounds, such as fluoboric acid, HF.BF 3 or 
 HBF 4 , fluosilicic acid, 2HF.SiF 4 or H 2 SiF 6 , chlorauric and 
 chloroplatinic acids, HCl.AuCl 3 or HAuCl 4 , 2HCl.PtCl 4 or 
 H 2 PtCl 6 , &c. 
 
 (b) Thio-acids, which may be regarded as oxygen acids, 
 a part or the whole of the oxygen of which is replaced by 
 sulphur, e.g. 
 
 Thiosulphuric acid . . . H 2 SO 3 S 
 Thiocarbonic acid . . . H 2 CS 3 
 Thiacetic acid . . . HC 2 H 3 OS 
 
 2. In proportion as the quantity of oxygen increases in 
 a series of acids containing the same elements, so the 
 acidity as well as the stability of the compound is almost 
 always increased. A detailed examination of such examples 
 as the following would, if space permitted, confirm this 
 statement : 
 
 HC1O HNO 2 H 2 SO 3 H 3 PO 2 
 
 HC10 2 HN0 3 H 2 S0 4 H 3 P0 3 
 
 HC10 3 H 3 PO 4 
 
 HC10 4 
 
 Glycolic acid, C 2 H 4 O 3 
 
 Oxalic acid, C 2 H 2 O 4 
 
 3. The addition of oxygen to (neutral) aldehyds converts 
 them into acids. Ex. gr. 
 
 Acetic. 
 
 Aldehyd . . . C 2 H 4 O 
 
 Acid .... C 2 H 4 O 2 
 
 Benzoic. 
 
 Aldehyd . . C 7 H 6 O 
 
 Acid .... C 7 H 6 O 2 
 
 4. In homologous series of acids, the lowest terms which 
 are richest in oxygen show a far stronger development of 
 
230 Chemical Philosophy. 
 
 the acid character than the higher terms of the same series, 
 which are comparatively poor in oxygen. Compare, for 
 example 
 
 Formic acid, CH 2 O 2 , containing 69-5 per cent, of oxygen, 
 Acetic acid, C 2 H 4 O 2 53-3 
 
 with 
 Valeric acid, C 5 H 10 O 2 31-3 
 
 and 
 Palmitic acid, C 16 H 32 O 2 12-4 
 
 5. Many oxides of metals are basic, that is, saturate 
 acids with production of water and salt, but a few examples 
 will be sufficient to show that it is only those which contain 
 a comparatively small quantity of oxygen that possess this 
 power. As the quantity of oxygen increases in a given 
 series, the basic character gradually disappears, and gives 
 place to a more or less decided acid-forming tendency. 
 
 EXAMPLES OF METALLIC OXIDES. 
 
 Basic 
 
 Hg 2 
 HgO 
 
 KjO 
 
 BaO 
 
 PbO 
 
 MnO 
 Mn 2 O 3 (feebly) 
 
 CrO 
 Cr A (feebly) 
 
 Interme- 
 diate 
 
 
 KA 
 
 K 2 4 
 
 Ba0 2 
 
 Pb0 2 
 
 MnO 2 
 
 
 Anhydric | 
 Acids \ 
 
 
 
 
 
 MnO 3 
 MnA 
 
 CrO s 
 
 Allotropic oxygen or ozone is a body which has attracted 
 considerable attention. The experiments of Brodie finally 
 decided the question of its constitution in favour of the 
 hypothesis long ago put forward by Odling. Ozone is now 
 proved to be allotropic oxygen, free from hydrogen, and to 
 have the formula O 3 . It may thus be regarded as formed 
 
Sulphur and Selenion. 231 
 
 on the type of hydric peroxide, with which body it agrees 
 in many of its reactions. 
 
 Hydric peroxide. Ozone. 
 
 H-0 /O 
 
 I O/ | 
 
 H-0 X 
 
 Sulphur, Selenion. These two elements resemble each 
 other closely. Sulphur is a yellow, selenion a red solid, 
 exhibiting several modifications, some of which are crystal- 
 line and soluble in carbon disulphide, others amorphous 
 and insoluble. 
 
 Principal Allotropes of Sulphur. 
 
 Rel. Dens. In carbon disulphide. 
 
 1. Octahedral (native) . 2-07 soluble. 
 
 2. Prismatic (monoclinic) rpS transformed into i. 
 
 3. Plastic . . .i'95 insoluble. 
 
 4. Amorphous . . . 1*95 insoluble. 
 (Precipitated from sulphur chloride or from thiosulphates.) 
 
 Principal Allotropes of Selenion. 
 
 1. Monoclinic (native) . 4*4 soluble. 
 
 2. Crystalline (form ?) . 4-8 insoluble. 
 
 3. Vitreous . . . 4*3 insoluble. 
 
 4. Amorphous ? insoluble. 
 
 (Precipitated from selenious acid.) 
 
 The following is a comparison of the most important 
 compounds of sulphur and selenion : 
 
 H,S 
 
 Gas. 
 
 S0 2 
 
 Gas. 
 
 S0 3 
 
 Solid. 
 
 H 2 S0 3 
 
 Crystallisable at 
 low temperatures. 
 
 H 2 SO 4 
 
 Oily liquid. 
 
 H 2 Se 
 
 Gas. 
 
 SeO 2 
 
 Solid. 
 
 Se0 3 
 
 Unknown. 
 
 H 2 Se0 3 
 
 Crystalline solid. 
 
 H 2 SeO 4 
 
 Oily liquid. 
 
 The selenates are isomorphous with the sulphates. 
 The relations of sulphur to oxygen are shown in the 
 following synopsis of some of their compounds : 
 
232 
 
 Chemical Philosophy. 
 
 SULPHUR. 
 
 OXYGEN. 
 
 SH 2 gas, feebly acid 
 S 2 H 2 oily liquid 
 CS red solid ? 
 CS 2 volatile liquid \ 
 COS gas 
 
 OH 2 
 O,H 2 oily liquid 
 CO colourless gas 
 CO 2 gas 
 
 Aci 
 
 C 2 H 3 O.SH thiacetic. B.P.93 
 SOyOH.SH thiosulphuric 
 
 DS. 
 
 C,H 3 O.OH acetic. B.P. 120 
 SO 2 .OH.OH sulphuric 
 
 BA< 
 
 SKH 
 
 SK 2 
 SCa 
 CS(NH 2 ) 2 thiocarbamide 
 
 ES. 
 
 OKH 
 
 OK 2 
 OCa 
 CO(NH 2 ) 2 carbamide or urea 
 
 SAI 
 
 M' 2 CS 3 thiocarbonate 
 M' 3 AsS 4 thioarsenate 
 M'CNS thiocyanate 
 
 .TS. 
 
 M' 2 CO 3 carbonate 
 M' 3 AsO 4 arsenate 
 M'CNO cyanate 
 
 ALCO 
 S { ^ HS mercaptan. B.P. 36 
 
 HOLS. 
 
 Q |QH 5 B p< 7 g . 4 
 
 ETt 
 
 S(C 2 H 5 ) 2 - B.P. 72 
 
 [ERS. 
 
 0(C 2 H 5 ) 2 . B.P. 35 -6 
 
 It is necessary, however, to add that many oxygen com- 
 pounds are known for which at present there are no corre- 
 sponding terms in the sulphur series the oxides of nitrogen, 
 for example. On the other hand, sulphur has a power of 
 accumulating in a manner which is not exhibited, at least 
 to the same extent, by oxygen. Consequently, many poly- 
 sulphides exist for which there are no corresponding oxides. 
 Thus we have K 2 S 5 , CaS 6 , FeS 2 , (C 2 H,) 2 S 2 , and (C 2 H 5 ) 2 S 3 . 
 
Water of Crystallisation. 233 
 
 Valency of Oxygen and Sulphur. 
 
 Oxygen is generally diad or bivalent, and in gaseous 
 compounds, with the exception of carbonic oxide, invariably 
 so. But that it is capable of acting the part of a tetrad 
 there can be no doubt, as in the oxides Cu 4 O and Ag 4 O. 
 Carbonic oxide is generally represented as containing un- 
 saturated carbon, C = O ; but admitting the essentially tetrad 
 character of the oxygen, we should represent it as constituted 
 thus, C=O. Again, water of crystallisation is almost uni- 
 versally considered by chemists to exist as water in the salts 
 in which it occurs, the molecules of the salt and the 
 molecules of the water retaining their individuality, being 
 united only by some kind of adhesion. Such water is very 
 easily detached, and constitutes no essential part of the 
 chemically reacting unit of the compound in all the most 
 characteristic of its transformations. Nevertheless, the idea 
 that this water is held to the salt by the two extra units of 
 combining power of the oxygen is quite worthy of con- 
 sideration. One example of the application of this hypo- 
 thesis will suffice. Zinc sulphate contains ZnSO 4 .7OH 2 , or 
 ZnSO 4 OH 2 .6OH 2 . This may be expressed in the following 
 manner : 
 
 z/ 
 \ 
 
 r 
 
 ) \. 
 \/ 
 
 s 
 
 /IK 
 \ / 
 
 j n 
 
 ^ TT 
 
 O 
 
 H., = O O = H. 
 
 = H 2 
 
 I I 
 H 2 = 0=H 2 
 
234 Chemical Philosophy. 
 
 Now it is well known that six out of the seven molecules 
 of water present in crystals of zinc sulphate are expelled by 
 heat much more readily than the last molecule, which can 
 be detached only with great difficulty. This fact is recorded 
 in the second of the two formulae given above, and it also 
 comes out in the graphic formula in which two atoms of 
 hydrogen are represented as occupying a position in the 
 molecule corresponding with that of the zinc. 
 
 With regard to sulphur, the question is somewhat different. 
 This element forms three well-marked classes of compounds. 
 In the first of these it is bivalent, and may be regarded as 
 the strict representative of oxygen. Examples of this type 
 of compound have been already given, SH 2 , S 2 H 2 , CS 2 , 
 COS. 
 
 In the second class the sulphur alone is quadrivalent : 
 ex. gr. in 
 
 Cl 
 
 * ( /V' 
 
 Cl and S O , 
 
 , 
 Cl 
 
 Sulphur tetrachloride. Thionyl chloride. Sulphur dioxide. 
 
 In the third class of sulphur compounds, represented by 
 sulphuric anhydride, SO 3 , the central atom seems to be sexi- 
 
 O 
 valent sulphur : || . It is true that the constitution 
 
 o=s=o 
 
 of this and all the rest of the sulphur compounds might 
 be represented on the hypothesis that sulphur is diad, 
 O S O 
 \ / , but the analogy of sulphur with chromium, and 
 
 O 
 
 especially the isomorphism of the sulphates and chromates, 
 seem to establish the title of sulphur to be regarded as a 
 hexad. 
 
Sulphur Compounds. 
 
 235 
 
 The several oxides, hydrates, and oxychlorides are there- 
 fore represented in the following manner : 
 
 ' 
 
 Sulphur trioxide. 
 
 O" 
 O" 
 Cl 
 
 Cl 
 
 Sulphuryl 
 chloride. 
 
 ( 0" 
 
 vi O" 
 
 S (OH)' 
 
 (b 
 
 ,0" 
 
 S P" 
 S (OH) 
 
 I (OH)' 
 
 Sulphuryl 
 chlorhydrate. 
 
 Sulphuryl hydrate, or 
 common sulphuric acid. 
 
 1O" 
 (OH)' 
 (OH) 
 (OH)' 
 (OH)' 
 
 Glacial or tetrahydric sulphuric 
 acid. 
 
 Ferrous sulphate 
 (dried at 100). 
 
 Orthosulphuric acid 
 (unknown). 
 
 (8>Hg 
 
 Trimercuric orthosulphate, 
 Turpeth mineral? 
 
 r S0 2 .OH 
 
 o 
 
 (S0 2 .OH 
 
 Pyrosulphuric or Nprdhausen 
 sulphuric acid. 
 
 fSO 2 .Cl 
 
 o 
 Iso 2 .ci 
 
 Pyrosulphuryl 
 chloride. 
 
 NON-METALS. CLASS 3 : BORON. 
 
 A triad element, but exhibiting marked analogies with 
 tetrad silicon. The following are their chief points of 
 agreement : 
 
Chemical Philosophy. 
 
 BORON. 
 
 Element, (a) amorphous brown 
 powder or () quadratic octa- 
 hedrons 
 
 Density 2-68 
 
 BH 3 inflammable gas 
 
 B(CH 3 ) 3 , B(C 2 H 5 ) 3 , &c. 
 BiCl 3 . B.P. 17 
 BF 3 gas 
 BF 3 .HForBHF 4 
 
 known only in solution : salts 
 
 crystallisable 
 H 3 B0 3 
 
 crystallisable 
 HB0 2 
 
 B,O 3 (fused, Dens. I -83) 
 M~' 3 BO 3 orthoborates 
 M'BO 2 metaborates 
 Many anhydroborates 
 M'BO 2 + *B 2 O 3 
 Ex. fused borax, 2NaBO 2 + B 2 O. 
 
 SILICON. 
 
 Element, (a} amorphous brown 
 powder or (b] hexagonal plates 
 
 Density about 2 '5 
 
 SiH 4 spontaneously inflammable 
 
 gas 
 
 Si(CH 3 ) 4 , Si(C 2 H 5 ) 4 , &c. 
 SiCl 4 . B.P. 59 
 SiF 4 gas 
 SiF 4 .2HF or SiH 2 F 6 
 
 scarcely known except in solu- 
 tion : salts tolerably stable 
 H 4 Si0 4 
 
 exists only in solution 
 H 2 SiO 3 
 
 Si(X (fused, Dens. 2'2) 
 M' 4 SiO 4 orthosilicates 
 M' 2 SiO 3 metasilicates 
 Many anhydrosilicates 
 M' 2 SiO 3 + *SiO 2 
 Ex. orthoclase felspar 
 K,,SiO 3 
 Al 2 (Si0 3 ) 3 
 
 Boron is one of the few elements which combine in a 
 direct manner with nitrogen. Boron nitride has the formula 
 BN, or perhaps B 3 N 3 . 
 
 NON-METALS. CLASS 4. 
 
 Carbon, C = 12. 
 
 Silicon, Si = 28. 
 
 Allotropes of Carbon. 
 
 a. Diamond, octahedral, Dens. 3-3 to 3*5. 
 
 b. Graphite, hexagonal, Dens, about 2-2. 
 
 c. Charcoal, amorphous. 
 
 Allotropes of Silicon. 
 
 a. Adamantine (?). 
 
 b. Graphitic, hexagonal, Dens, about 2 '5. 
 
 c. Amorphous. 
 
Carbon compared with Silicon. 237 
 
 Carbon is remarkable as being the essential element of 
 organic nature, and silicon as being one of the most 
 abundant constituents of the earth's solid crust. Regarded 
 as chemical elements, they present problems of special 
 interest in the multitudinous array of compounds which 
 they have furnished to the chemist. The carbon com- 
 pounds will be specially considered hereafter ; it therefore 
 only remains to point out in this place a few of the remark- 
 able coincidences of composition so frequently observed 
 between the compounds of these two elements. The 
 number of bodies in which silicon replaces one or more 
 atoms of the carbon in some already familiar compound is 
 rapidly increasing, and still further developments in this 
 direction may be looked for. 
 
 In comparing carbon and silicon compounds together, it 
 cannot be said that the similarity so often noticed in their 
 constitution extends to their properties. In most cases 
 corresponding carbon and silicon compounds are very 
 different bodies. A few examples only can be referred to : 
 
 Carbon Compounds. 
 CO, colourless inflammable gas. 
 CO 2 , colourless gas, soluble to 
 some extent in water. 
 
 Silicon Compounds. 
 
 SiO, unknown. 
 
 SiO 2 , crystalline solid, fusible only 
 at a very high temperature, 
 totally insoluble in water. Hy- 
 drates slightly soluble, 
 
 SiH 4 , spontaneously inflammable 
 
 gas. 
 
 water. 
 
 liquid, decomposed by 
 
 CH 4 , permanent gas. 
 
 CHC1 3 , heavy liquid, unacted upon 
 by water. 
 
 Orthosilicicand orthocarbonic acids, H 4 SiO 4 and H 4 CO 4 , 
 are unknown, and but very few of their compounds have 
 been studied. Of orthosilicates we have olivine, Mg // 2 SiO 4 , 
 ferrous orthosilicate, Fe // 2 SiO 4 , the ethyl salt, (C 2 H 5 ) 4 SiO 4 , 
 and a few others. Orthocarbonates are represented by such 
 compounds as the ammonium-hydrogen orthocarbonate 
 (NH 4 ) 2 H 2 CO 4 , and the ether (C 2 H 5 ) 4 CO 4 . Ordinary car- 
 bonates are derived from metacarbonic acid, H 2 CO 3 , a body 
 which has not been isolated, though the corresponding 
 
238 Chemical Philosophy. 
 
 silicic acid has been obtained as a solid glassy mass. All 
 carbonates are decomposed on the addition of even the 
 weakest acids, whereas the majority of the silicates are 
 attacked but very slowly by acids. Nevertheless silicates 
 containing alkaline or earthy bases are slowly decomposed 
 even by carbonic acid, which extracts the alkali, leaving 
 hydrated silica, alumina and ferric oxide, &c. 
 
 Valency of Carbon and Silicon. 
 
 Silicon is always tetrad. Carbon is tetrad in all its known 
 compounds, except possibly carbonic oxide, CO, and the 
 isocyanides. The unsaturated character of carbonic oxide 
 is shown by the readiness with which it enters into combi- 
 nation with chlorine, with oxygen, with potassium, with 
 caustic potash, and with cuprous or platinous chloride. 
 
 CO + C1 2 = COC1 2 , carbonyl chloride 
 CO + O = CO 2 , carbon dioxide 
 CO + K = K n C M O n 
 CO + KHO = KCHO 2 , formate, &c. 
 
 From the characteristic reactions of the isocyanides (see 
 Isomerism, Chap. XXI., p. 197) it appears that in these 
 bodies an atom of carbon is linked to the hydrogen, or other 
 positive radicle present, by the atom of nitrogen. Now, ac- 
 cording as we regard the nitrogen in these compounds as 
 discharging a triad or pentad function, we must attribute to 
 the carbon a replacing value equal to two or four atoms 
 of hydrogen. Either of these two formulae, then, may be 
 adopted : 
 
 // /// / iv v / 
 
 C = N R or CEEN R. 
 
 The one may be regarded as a derivative of ammonia, the 
 other of marsh-gas. 
 
 In favour of the former view we may adduce the combi- 
 nation of such a cyanide as prussic acid with hydrochloric, 
 
Nitrogen and Phosphorus. 239 
 
 hydrobromic, and hydriodic acids, forming crystalline com- 
 pounds analogous to sal-ammoniac, and with hydrogen form- 
 ing methylamine \ also the combination of the isocyanides 
 generally with oxygen to form isocyanates, and with sulphur 
 to form isothiocyanates. These combinations may be 
 represented thus : 
 
 XT (H XT (H XT (H XT (H vr|H 2 
 
 N |C" N |C" N |c = N {C = S N 1CH 3 
 HC1 
 
 Hydrocyanic Carbylamine Cyanic acid. Thiocyanic Methylamine. 
 
 acid or hydrochloride. acid. 
 
 Carbylamine. 
 
 If the second hypothesis is adopted, we must imagine 
 that the union of the carbon to the nitrogen is partly un- 
 locked before these compounds can be found. Written on 
 the marsh-gas type the formulae then become 
 
 NON-METALS. CLASS 5 I NITROGEN AND PHOSPHORUS. 
 
 Nitrogen N = I4) N , V Phosphorus? = 31 )p,q h 
 
 Phosphorus P = 3i [ = 32- $ Arsenic As = 75 [ -=75'5 
 Vanadium V = 5 1 J 2 Antimony Sb = 1 20 ) 2 
 
 Phosphorus P = 3 1 \ p As Phosphorus P = 3 1 ) p , TK 
 
 Vanadium V*> 51 [ = 53 Antimony Sb = 1 20 I = 1 1 9 -5 
 
 Arsenic As = 75) 2 Bismuth Bi = 2o8J 2 
 
 Diagram of Numerical Relations typical of Chemical 
 
 Relations. 
 Ni 4 /V 5 i-As 7 5 
 
 P3i As75 Sbi2o 
 
 i \ 
 
 51 \Sbi2o-Bi2o8 
 
 Nitrogen is a colourless gas, very slightly soluble in water, 
 liquefiable with great difficulty, and not known in any allo- 
 tropic form. 
 
 Phosphorus, on the contrary, is, in its ordinary condi- 
 
240 Chemical Philosophy. 
 
 tion, a crystalline solid, though capable of passing into several 
 allotropic modifications. These have already been referred 
 to (Chap. XXL, p. 199). 
 
 The relative density of nitrogen agrees with the mole- 
 cular formula N 2 , whilst the vapour density of phosphorus 
 corresponds with P 4 . This peculiarity of phosphorus is 
 shared by the semi-metal arsenic. 
 
 Both nitrogen and phosphorus are intimately concerned 
 in the processes of animal life, phosphorus being especially 
 abundant in the form of phosphates in bone, and in nervous 
 tissue. 
 
 Nitrogen is connected with phosphorus mainly by reason 
 of the resemblance between ammonia, NH 3 , and phosphine, 
 PH 3 , and the bases formed from them by substitution of 
 hydrocarbon radicles for the hydrogen. The element itself, 
 as a permanent incombustible gas, is wholly unlike the 
 solid inflammable phosphorus, whilst its oxides and acids, 
 in their volatility and easy decomposability, form a strong 
 contrast to the corresponding compounds of phosphorus. 
 The accident of agreement in their crude or synoptic 
 formulae by no means implies identity of constitution, and, 
 for the reasons just alluded to, it seems not improbable that 
 the atomic structure of the nitrogen oxides and acids is, at 
 least in many cases, different from that of the phosphorus 
 compounds. 
 
 The following are almost the only known examples of 
 concordance in the crude formulae of the oxygen com- 
 pounds : 
 
 Nitrogen Phosphorus 
 
 trioxide. trioxide. 
 
 N 2 3 P 2 3 
 
 Tetroxide. Tetroxide. 
 
 N 2 4 P 2 4 
 
 Pentoxide. Pentoxide. 
 
 N 2 5 P 2 5 
 
 Nitric acid. Metaphosphoric acid. 
 
 HN0 3 HPO 3 
 
Nitrogen and its Compounds. 241 
 
 The compounds N 2 O, NO, HNO 2 , find no parallels 
 in the phosphorus series, whilst H 3 PO 4 , H 3 PO 3 , H 3 PO 2 , 
 H 4 P 2 O 7 , the salts of which, at least, are highly stable and 
 definite, are altogether unrepresented arrlong the nitrogen 
 compounds. Again, whilst phosphorus yields two chlo- 
 rides, PC1 3 and PC1 5 , and an oxychloride, POC1 3 , nitrogen 
 forms no chloride, 1 but an oxychloride, NOC1, a yellow 
 gas. It has been the very general custom to represent all 
 the compounds of nitrogen by formulae framed in the same 
 manner as those of the phosphoric compounds, without 
 regard to their dissimilar properties. But by assuming that 
 nitrogen is trivalent in these bodies we render some account 
 of these discrepancies. On this view the formulae given 
 above for. nitric acid and nitric anhydride may be expanded 
 in the following manner : 2 
 
 Nitric pentoxide. Phosphoric pentoxide. 
 
 O, /O 0=P O P=0 
 
 I \ N _0-N/ | |i || 
 
 O/ X) 00 
 
 Nitric acid. Metaphosphoric acid. 
 
 O v O=P O H 
 
 | \N-0-H || 
 
 O/ O 
 
 In the ammonium compounds, on the other hand 
 (p. 308), nitrogen is probably pentad chloride of ammo- 
 nium, for instance, being N V H 4 C1. 
 
 Phosphorus. The two chlorides PC1 3 and PC1 5 may 
 be taken as the types of the two chief classes of compounds 
 which phosphorus is capable of forming. 
 
 1 The so-called chloride of nitrogen, obtained by the action of 
 chlorine upon sal-ammoniac, probably contains hydrogen. Its formula 
 is supposed to be NHC1 2 , but in consequence of its dangerously ex- 
 plosive properties it has been only imperfectly examined. 
 
 2 The formulae might also be written O = N-O-O-O-N = O, and 
 O = N O O H, though for several reasons the form given in the 
 text is preferable. 
 
 R 
 
242 Chemical Philosophy. 
 
 The trichloride PC1 3 is a colourless volatile liquid, which 
 combines in a direct manner with chlorine, with oxygen, or 
 with sulphur, yielding the pentachloride, PC1. 5 , the oxy- 
 chloride, PC1 3 O, and the sulphochloride, PC1 3 S, in which 
 compounds the phosphorus appears to be saturated. 
 
 The best known iodide is anomalous. It has the for- 
 mula P 2 I 4 , and corresponds with a liquid hydrogen com- 
 pound P 2 H 4 , to which the spontaneous inflammability of 
 phosphoretted hydrogen is stated to be due. 
 
 Phosphine, or phosphoretted hydrogen, PH 3 , is a highly 
 inflammable gas, which so far imitates ammonia ] that it 
 combines with hydriodic acid to form a salt, phosphonium 
 iodide, PH 4 I, crystallisable in cubes. 
 
 The compounds in which phosphorus plays the part of 
 a pentad are more numerous. Besides the pentachloride 
 PC1 6 , and pentoxide P 2 O 5 , and intermediate PC1 3 O, we 
 have the following series of oxidised bodies which may 
 be considered as derived from phosphine, either by addition 
 of oxygen or by substitution of hydroxyl for its hydrogen: 
 
 H 
 
 Phosphine. / 
 
 H 3 P or P-H 
 
 \ 
 
 H 
 
 Phosphine oxide. 
 
 H 8 PO, unknown, but represented by 
 
 C 2 H 5 
 
 Triethyl -phosphine oxide. / 
 
 (C 2 H 5 ) 3 PO or O=P C 2 H 5 
 
 C 2 H 5 
 OH 
 
 Hypophosphorous acid (monobasic). / 
 
 H 3 PO 2 or O=P H 
 
 H 
 
 1 See also Amines, p. 310. 
 
Acids containing Phosphorus. 243 
 
 OH 
 
 Phosphorous acid (dibasic). / 
 
 H 3 P0 3 or 0=P OH 
 
 H 
 OH 
 
 Orthophosphoric acid (tribasic). / 
 
 H 3 PO 4 or O=P-OH 
 
 OH 
 
 OH 
 
 Pyrophosphoric acid (tetrabasic). / 
 
 H 4 P 2 O 7 or O=P 
 
 NDH 
 
 t, 
 
 x 
 
 0=P 
 
 X O 
 
 Metaphosphoric acid (monobasic). ^ 
 
 HPO 3 or O=P 
 
 N)H 
 
 With the exception of the first two compounds in the 
 list, phosphine and its oxide, these bodies are all powerful 
 acids, syrupy, glassy, or crystalline, according to circum- 
 stances. They cannot be made to evolve oxygen at any 
 temperature as nitrous and nitric acids do. Under the 
 influence of heat, hypophosphorous and phosphorous acids 
 emit phosphine, leaving a residue of phosphoric acid 
 
 2 H 3 P0 2 = H 3 P0 4 + H 3 P 
 and 4 H 3 P0 3 = 3 H 3 P0 4 + H 3 P; 
 
 whilst orthophosphoric acid by the same treatment loses 
 
 R 2 
 
244 Chemical Philosophy. 
 
 water, though the dehydration can by no means be carried 
 beyond the production of metaphosphoric acid : 
 
 2 H 3 P0 4 - H 2 = H 4 P 2 7 
 and H 3 PO 4 - H 9 O = HPO 3 
 or H 4 P 2 7 - H 2 = 2 HP0 3 
 
 By boiling either of these products with water the ortho- 
 acid is re-generated. 
 
 Metaphosphoric acid is convertible, by the action of heat, 
 into several polymeric acids, H w P n O 3n . 
 
 Phosphorus is closely connected with the semi-metals 
 vanadium and arsenic, by reason of the very general iso- 
 morphism of the phosphates with the arsenates and vana- 
 dates. This relationship will be adverted to in a subsequent 
 chapter. 
 
 CHAPTER XXIV. 
 DIVISION II. METALLOIDS. 
 
 THE name metalloid obviously belongs to bodies which have 
 the form or appearance of a metal, or which in some way 
 resemble a metal. Hence the impropriety of applying this 
 term, as is not uncommonly done, to such elements as the 
 halogens, to nitrogen, or phosphorus. The name is in- 
 tended in this volume to designate a body which, although 
 resembling a metal in most characteristics, yet lacks some 
 one or more of the features which true metals generally 
 present. It applies to a somewhat miscellaneous set of 
 bodies. The question of the propriety of including hydro- 
 gen amongst them will be discussed presently, but, leaving 
 this one element out of consideration, the test by which the 
 sub-metallic character of most of them may be detected is 
 the formation of more or less definite oxy-salts. Their oxides 
 are for the most part well-marked anhydric acids, but each 
 of them is capable of producing at least one oxide which 
 
Hydrogen. 245 
 
 possesses basic tendencies more or less pronounced. Still, 
 putting hydrogen aside, the metalloids are comparatively 
 imperfect conductors of heat and electricity, and generally 
 brittle. Tin is, however, both malleable and moderately 
 ductile. 
 
 METALLOIDS. I. HYDROGEN. 
 
 Hydrogen is isolated from all other known bodies by 
 reason of its extreme lightness, its relative density being 
 only "0693 when air is i. Chemically it differs from all the 
 non-metals in manifesting no tendency to combine with 
 metals ; only one such compound, cuprous hydride, Cu 2 H 2 , 
 being known. 1 On the other hand, hydrogen is, of all the 
 elements, most ready to lend itself to that peculiar state of 
 mechanical combination with metals known as * occlusion ' 
 (Chap. VI., p. 44). Palladium, for example, absorbs hydro- 
 gen largely, and the charged metal is indistinguishable in 
 appearance from pure palladium, though its conducting 
 power is slightly diminished. This body was regarded by 
 Graham, who discovered it, as the realisation of the long- 
 cherished idea of the essentially metallic character of hydro- 
 gen. But its title to recognition as a metal or metalloid is 
 based more securely upon the resemblance which acids or 
 hydrogen salts bear to metallic salts. In spite of a consider- 
 able number of gaseous and liquid exceptions, the majority 
 of known acids are crystallisable bodies, often combining 
 with water of crystallisation, and utterly indistinguishable to 
 all appearance from ordinary saline compounds. Among 
 common acids, for example, the following are crystallis- 
 able : 
 
 Sulphuric acid . . H 2 SO 4 (below 35).' 
 Glacial sulphuric acid . H 4 SO 5 (below 9). 
 Phosphoric acid . . H 3 PO 4 . 
 
 1 Definite compounds of hydrogen with palladium, sodium, and 
 potassium, having the formulae Pd 2 H, Na 2 H, and K 2 H, are said to be 
 produced by heating these metals in hydrogen gas. It is, however, not 
 yet certain that these are true chemical compounds. 
 
246 Chemical Philosophy. 
 
 Boracic acid . . . H 3 BO 3 . 
 
 Oxalic acid . . . H 2 C 2 O 4 + 2H 2 O. 
 
 Formic acid. . . HCHO 2 (below i). 
 
 Acetic acid . . . HC 2 H 3 O 2 (below 17). 
 
 Tartaricacid . . H 2 C 4 H 4 O 6 . 
 
 Citric acid . . . H 3 C 6 H 5 O 7 + H 2 O. 
 
 Gallic acid . . . HC y H 5 O 5 + H 2 O. 
 
 But the resemblance between hydrogen salts and metallic 
 salts is not limited to external form. They agree in chemical 
 reactions. When an acid is acted upon by a metal, and 
 hydrogen is expelled, the metathesis is entirely comparable 
 with the exchange of a more positive for a less positive 
 metal of the ordinary recognised class. For example, zinc 
 decomposes hydrogen sulphate, just as it decomposes copper 
 sulphate ; in each case the less positive metal is displaced, 
 and a new salt formed. 
 
 ]H 2 SO 4 + Zn = H 2 + ZnSO 4 . 
 (CuSO 4 -f Zn = Cu + ZnSO 4 . 
 
 Again, water that is, hydrogen oxide or hydroxide, H 2 O 
 or H(HO) not unfrequently imitates metallic oxides or 
 hydroxides in their chemical reactions. In order to produce 
 exactly the same effect in any given case, it is only necessary 
 to use the water in larger quantity or at a higher tempera- 
 ture. 
 
 The following are examples of this mode of action : 
 MgCl 2 + HgO = MgO + HgCl 2 . 
 MgCl 2 + H 2 O = MgO + 2 HC1. 
 Fe 2 Cl 6 + 6NaHO = Fe 2 (HO) 6 + 6NaCl. 
 Fe 2 Cl 6 + 6HHO = Fe 2 (HO) 6 + 6HC1. 
 
 r(C 3 H 5 /"Sl 3 1 + 3 NaHO = Q,H 5 (HO) 3 + 3 NaSt. 
 
 Stearine. Glycerine. Sodium 
 
 stearate. 
 
 (C 3 H 5 )'"St 3 + 3 HHO = C 3 H 5 (HO) 3 -f 3 HSt 
 C 18 H 35 2 . 
 
 Stearine. Glycerine. Hydrogen stearate, 
 
 or stearic acid. 
 
Tellurium. 247 
 
 Water and acids also undergo electrolysis in the same 
 manner as metallic salts ; e.g. 
 
 2 HHO = H 2 + (HO)a 1 
 2 HC1 = H 2 + C1 2 
 CuCl 2 = Cu + C1 2 
 
 Lastly, it might be pointed out that the sour taste of 
 hydrogen salts and their reddening effect upon vegetable 
 blues are no more to be regarded as indications of peculiar 
 constitution than is the alkaline reaction of many normal 
 metallic salts or the neutrality of others. In such a series 
 as the following we see a gradual transition from strongly 
 alkaline to strongly acid properties : 
 
 Na 3 PO 4 Na 2 HPO 4 NaH 2 PO 4 H 3 PO 4 
 
 Very alkaline. Alkaline. Acid. Very acid. 
 
 Enough has been said to indicate to the student that, 
 despite many peculiarities which are in all probability con- 
 nected with its remarkable physical characters, hydrogen 
 displays, from a chemical point of view, features bearing an 
 unmistakable metallic impress. 
 
 METALLOIDS. 2. TELLURIUM. 
 
 Tellurium in many respects closely imitates sulphur and 
 selenion. Thus it forms a gaseous telluretted hydrogen, 
 H 2 Te, the representative of hydrogen sulphide ; also oxides 
 TeO 2 and TeO 3 , and acids H 2 TeO 3 and H 2 TeO 4 , corre- 
 sponding respectively with sulphur dioxide and trioxide, and 
 with sulphurous and sulphuric acids. But tellurium is 
 decidedly metallic in appearance, is thrown down from 
 solution in the form of sulphide by sulphuretted hydrogen, 
 
 1 It is a question whether pure water is an electrolyte at all, but 
 supposing it to be capable of electrolysis this is probably the first phase 
 in the decomposition, the oxygen which is evolved being due to a 
 secondary reaction, thus : 
 
 2HHO = H 2 + (HO), = H 2 + H 2 O + O. 
 
248 
 
 Chemical Philosophy. 
 
 and its lower oxide saturates acids forming salts, of which 
 the sulphate, Te(SO 4 ) 2 , and nitrate, Te(NO 3 ) 4 , are sufficient 
 examples. 
 
 METALLOIDS. 3. GERMANIUM, TIN, TITANIUM, ZIRCONIUM. 
 
 Save for the distinctly metallic character of elemental tin 
 and germanium, and production by tin, germanium, and 
 titanium of an inferior diadic class of derivatives, these ele- 
 ments bear a strong family resemblance to silicon. That the 
 metallic function is more or less distinctly developed in them, 
 whilst it is altogether imperceptible in silicon, is also indicated 
 by the existence of various oxy-salts, sulphates, phosphates, 
 &c., of tin, titanium, and zirconium, and probably of germa- 
 nium, similar compounds of silicon being entirely unknown. 
 
 The most characteristic compounds of these elements 
 are the chlorides, oxides, and acids in which they play the 
 part of tetrads. The following are some examples compared 
 with corresponding silicon compounds : 
 
 Silicon 
 Si =28 
 
 Germanium 
 
 Tin 
 Sn = 118 
 
 Titanium 
 Ti = 4 8 
 
 Zirconium 
 Zr=Sg-6 
 
 SiCl 4 
 
 liquid, B.P. 
 50 
 
 GeCl 4 
 
 liquid, B.P. 
 87 
 
 SnCl 4 
 liquid, B.P. 
 
 120 
 
 TiCl 4 
 
 liquid, B.P. 
 135 
 
 ZrCl 4 
 
 volatile solid 
 
 SiF 4 
 
 liquefiable 
 gas 
 
 GeF 4 
 
 solid? 
 
 SnF 4 
 
 fuming liquid? 
 
 TiF 4 
 
 fuming liquid 
 
 ZrF 4 
 
 solid 
 
 M' 2 SiF 6 
 Fluosilicates 
 
 M' 2 GeF 6 
 Fluo- 
 germanates 
 
 M',SnF 6 
 Fluostannates 
 
 M' 2 TiF B 
 
 Fluotitanates 
 
 Fluozirconates 
 
 Corresponding salts isomorphous. 
 
 SiO 2 | Ge0 2 
 
 Sn0 2 
 
 Ti0 2 
 
 Zr0 2 
 
 Isomorphous 
 
 in tinstone. in rutile. artificial. 
 
Elements allied with Tin. 249 
 
 Valency of Tin and Allied Elements. 
 
 As already mentioned, tin forms, in addition to the well- 
 defined tetrad compounds, some of which are represented 
 in the foregoing table, an oxide, Sn n O n , a chloride, Sn n Cl 2n , 
 a sulphide, Sn n S M , and some other compounds in which 
 it seems to be diad. In the absence of direct evidence 
 in favour of the one view or the other, it is customary 
 to represent the oxide as SnO, and the chloride as SnCl 2 . 
 In favour of the formula SnO we have its analogy to that 
 of carbonic oxide, CO, though there is absolutely no resem- 
 blance in the properties of the two bodies to support such a 
 view. 
 
 For stannous chloride, representing it as a dichloride, 
 we have no example among the compounds of carbon and 
 silicon. On the whole it seems, therefore, probable that 
 the tin retains its quadrivalence in both the oxide and the 
 chloride, and their formulae in accordance with this view 
 would be written thus : 
 
 Sn 2 Cl 4 or C1 2 = Sn = Sn = C1 2 
 Sn 2 O 2 or O = Sn = Sn = O. 
 
 The element germanium was discovered in 1886 in a 
 rare mineral, argyrodite, 3Ag 2 S.GeS 2 , found at Freiberg. It 
 is brittle and volatile, and crystallises in octahedrons of 
 density 5-47, and melting at about 900. It forms two classes 
 of compounds somewhat similar to those of tin, including a 
 yellow monosulphide, GeS, and a white disulphide, GeS 2 . 
 
 It also forms a tetrethide, Ge(C 2 H 5 ) 4 , which boils at 
 1 60, but no compound with hydrogen has yet been ob- 
 tained. 
 
250 Chemical Philosophy. 
 
 METALLOIDS. 4. VANADIUM, ARSENIC, ANTIMONY, BISMUTH. 
 
 (For numerical relations of atomic weights see Non-metals, 
 Class 5, p. 239.) 
 
 These elements are all quinquivalent in their most 
 advanced states of combination, the tendency to assume 
 this condition being the least prominent in the case ot 
 bismuth, which is of the whole class the most decidedly 
 metallic. 
 
 Vanadium is remarkable for the extensive series of 
 oxides, chlorides, and oxychlorides to which it gives rise, 
 the different stages of oxidation of its dissolved compounds 
 being characterised by a strange variety of colour, a yellow 
 or red tint belonging to the higher, and a blue or green tint 
 especially to the lower oxides, which thus resemble some 
 of the compounds of chromium, molybdenum, and tungsten. 
 Oxides of phosphorus, arsenic, antimony, and bismuth, ex- 
 cept Bi 2 O 5 , are white. 
 
 The known oxides of vanadium have the formulae V 2 O, 
 VO or V 2 O 2 , V 2 O 3 , VO 2 or V 2 O 4 , and V 2 O 5 , thus running 
 parallel with the series of nitrogen oxides. There are three 
 chlorides, VC1 2 or V 2 C1 4 , VC1 3 , and VC1 4 . The last is a 
 brown liquid boiling at 154, and yielding a vapour of 
 density 96*5, which is half the molecular weight represented 
 by the formula VC1 4 . Vanadium, therefore, does not main- 
 tain the usual persistence of odd or even valency through- 
 out its combinations (see Chap. XVI.). 
 
 The vanadates, M' 3 VO 4 , are isomorphous with the phos- 
 phates, M' 3 PO 4 . Metavanadates and pyrovanadates also 
 exist, corresponding with meta- and pyro-phosphates, besides 
 a certain number of anhydro-meta-salts, M"(VO 3 ) 2 .V 2 O 5 and 
 M''(V0 3 ) 2 .2V 2 5 . 
 
 The analogies of arsenic, antimony, and bismuth will 
 be most satisfactorily brought out by tabulating the formulae 
 
The Arsenic Group. 
 
 251 
 
 of their leading compounds. For the sake of comparison 
 the corresponding compounds of phosphorus are placed side 
 by side with them. 
 
 The vapour densities of phosphorus and arsenic lead to 
 the molecular formulae P 4 and As 4 for those elements ; but 
 antimony and bismuth, though easily fusible metals, are not 
 volatile enough to give vapour at any manageable tempera- 
 ture. Their molecular formulae are therefore unknown. 
 
 Phosphorus 
 
 Arsenic 
 
 Antimony 
 
 Bismuth 
 
 
 Hydrides and Ethylides. 
 
 
 P ( cf Ha ) 3 
 
 AsH 3 
 
 As(C 2 H 5 ) 3 
 
 SbH, ] 
 Sb(C 2 H 3 ) 3 | Bi(C 2 H 5 ) 3 
 
 Chlorides. 
 
 PCI, 
 PC1 5 
 
 AsCl 3 
 
 SbCl 3 
 SbCl 5 
 
 , BiCl 3 
 
 Oxides. 
 
 PA 
 
 (As 2 3 ) 2 
 
 (Sb 2 O 3 ) 2 
 Sb^O 4 
 
 Bi,O 3 
 Bi."O 4 
 
 Acids. 
 
 HPH 2 2 
 
 H,PH0 3 
 I IPO, 
 H :i P0 4 
 
 HAsOj ? 
 IIjAsO. 
 HAsO, 
 
 H,As0 4 
 H 4 AsA 
 
 HSbO 2 
 
 HSb0 3 
 II 3 Sb0 4 ? 
 H 4 Sb,0 7 
 
 HBiO 2 
 HBiO 3 
 
 Sulphides. 
 
 P 2 S 
 P,S 3 
 
 1:1: 
 
 Sb,S 3 
 
 Bi 2 S 2 
 Bi 2 S 3 
 
 A few peculiarities exhibited by particular compounds 
 occurring in this table demand notice. 
 
252 Chemical Philosophy. 
 
 1. The faculty of uniting with acids possessed in so re- 
 markable a manner by ammonia is displayed by phos- 
 phoretted hydrogen towards hydnodic acid only, the corre- 
 sponding compounds with hydrobromic and hydrochloric 
 acids being produced only by compressing the mixed gases. 
 Arsenic and antimony hydrides do not combine with acids. 
 
 2. The ethylated and corresponding methylated com- 
 pounds are characterised not so much by a power of 
 combining with acids as by a tendency to unite with O, S, 
 Clj, Ij, &c, whereby the central element is brought to a 
 state of saturation. 
 
 3. No pentachloride, pentabromide, oxychloride, or oxy- 
 bromide of arsenic is known. 
 
 4. The vapour density of arsenious oxide (white arsenic) 
 is 198. Hence its molecular weight is 198 x 2 = 396, and 
 formula As 4 O 6 . It is usual to represent the oxides of 
 phosphorus by the formulae written in the table ; but if the 
 trioxide could be obtained in a pure state and its vapour 
 density determined, it is by no means improbable that its 
 molecular formula would also be doubled. 
 
 White arsenic occurs in commerce in the form of white 
 masses with conchoidal fracture, often exhibiting a banded 
 appearance, due to the presence of the isomeric vitreous 
 and porcellanous varieties. The former is converted into 
 the latter by pulverisation or by protracted boiling with 
 water. Either variety is but slightly soluble in water, and 
 the solution gives only a feebly acid reaction. Porcellanous 
 arsenic probably owes its opacity to crystalline structure. 
 By careful sublimation at regulated temperatures white 
 arsenic may be obtained either in the form of transparent 
 regular octahedra or in prismatic crystals of the trimetric 
 system, It is, therefore, isodimorphous with antimonious 
 oxide. 
 
 White arsenic is a feebly basic, as well as acid, anhy- 
 dride. Its best known salt is the double tartrate, 
 K(AsO)C 4 H 4 O 6 , which is analogous to the potassio-tartrates 
 
TJie Arsenic Group. 253 
 
 of antimony, K(SbO)C 4 H 4 O 6 (tartar emetic), and boron, 
 
 5. Phosphoric oxide is volatile and stable, but 
 Sb 2 O 5 , BigOj, are resolved by heat into oxygen and a lower 
 oxide. 
 
 6. Arsenious acid appears to be normally tribasic, the 
 yellow silver salt containing Ag 3 AsO 3 . It therefore differs 
 from phosphorous acid, which is dibasic. 
 
 Phosphorous acid Aisenioos acid. 
 
 OH OH 
 
 O = P OH As OH 
 
 \ \ 
 
 H OH 
 
 7. The orthophosphates and arsenates correspond in 
 every respect with each other, and in many cases with the 
 orthovanadates. In addition to the ordinary salts of the 
 common phosphate type, M'HsPO^ M'jHPO^ and M' 3 PO 4 , 
 there is a remarkable series of compounds occurring as 
 minerals in combination with a chloride or fluoride. 
 
 Apatite 3[Ca 3 (PO 4 ),] . CaF 8 or Ca 5 (PO 4 ) 3 F 
 
 or 3 [Ca 3 (P0 4 ) 2 ] .CaCl, or Ca 5 (PO 4 ) 3 Cl 
 Pyromorphite 3 [Pb 3 (P0 4 ) J .PbCl a or Pb 5 (PO 4 ) 3 Cl 
 Mimetine 3[Pb 3 (AsO 4 )J . PbCl, or Pb 5 (AsO 4 ) 3 a 
 Vanadite 3[Pb 3 (VO 4 ) 2 ] . PbCl, or Pb s (VO 4 ) 3 Cl. 
 
 Orthoantimonates and orthobismuthates are unknown, un- 
 we consider the intermediate tetroxides in that light: 
 antimonious orthoantimonate, Sb,O 4 or Sb"'(Sb T O 4 ), and 
 bismuthous orthobismuthate, Bi a O 4 or Br"'(Bi T O 4 ). Several 
 orthothioamimonates are, however, known ; the sodium 
 salt, for example, Na 3 SbS 4 .9H 2 O, crystallising readily. 
 
 8. The sulphides of arsenic, antimony, and bismuth act 
 as sulphur anhydrides, and in combination with the sul- 
 phides of silver, lead, copper, and other metals constitute 
 several interesting minerals. 
 
254 Chemical Philosophy. 
 
 METALLOIDS. 5. NIOBIUM, TANTALUM. 
 
 These two rare metals may be regarded as about equally 
 related on the one hand to tetrad tin and its allies, and on 
 the other to hexad molybdenum and tungsten. With the 
 latter elements, indeed, there is not only considerable 
 analogy of properties, but a curious approximation of 
 atomic weights. 
 
 Mo vi 96 Nb v 94 
 
 W vi 184 Ta v 182 
 
 Niobium and tantalum are pentad elements, and are 
 characterised by the production of pentachlorides in the 
 form of yellow, fusible, volatile bodies, the vapour densities 
 of which accord with the formulae NbCl 5 and TaCl 5 re- 
 spectively. 
 
 The metals themselves are only known in the condition 
 of black powders, which resist the attacks of most acids, 
 except hydrofluoric acid, in which they are slowly soluble 
 with evolution of hydrogen. Their fluorides unite with 
 the fluorides of the alkali-metals, forming soluble and crys- 
 tallisable double salts, 2KF,NbF 5 and 2KF,TaF 5 . But 
 niobium, at least, is also specially characterised by the 
 production of oxyfluorides, which, in combination with the 
 fluorides of the more positive metals, form salts which are 
 isomorphous with the fluotitanates, fluostannates, fluotung- 
 states, and fluozirconates. The following pairs of com- 
 pounds, for example, crystallise in the same form : 
 
 3 KF . HF . NbOF 3 
 3KF . HF . SnF 4 
 
 2 NH 4 F . NbOF 3 
 2NH 4 F . WO 2 F 2 
 
 3 NH 4 F . NbOF 3 
 3 NH 4 F . ZrF 4 
 
 2 NH 4 F . CuWO 2 F 4 
 2NH 4 F . CuTiF 6 
 
Molybdenum , Tungsten, Uranium. 255 
 
 It thus appears that groups of elements of different 
 valencies may occasionally replace one another isomor- 
 phously, provided the valency or atomicity of the various 
 groups is the same. 
 
 The radicles which are thus at the same time chemically 
 and isomorphously equivalent to one another in the fore- 
 going compounds are as follows : 
 
 (TiF 2 )", (ZrF 2 )", (SnF 2 )", (NbOF)", and (WO 2 )". 
 
 Such relations are of precisely the same character as 
 are those of the group (NH 4 )', ammonium, to the single 
 metallic atom K', and precisely the same conclusions 
 may be deduced from them. 
 
 METALLOIDS. 6. 
 
 Molybdenum Mo = 957 
 Tungsten W =184 
 
 Uranium U = 239. 
 
 These elements are heavily metallic bodies, tungsten and 
 uranium being especially distinguished both by high atomic 
 weight and specific gravity. 
 
 They appear to be hexads, though, of the three, tungsten 
 alone produces a hexchloride, and that unstable. In the 
 formation of trioxides and volatile dioxydichlorides (general 
 formula M vi O. 2 Cl 2 ) they resemble chromium, but they differ 
 from that metal in furnishing neither sesquioxide nor mon- 
 oxide analogous to Cr 2 O 3 and CrO. 
 
 The infraction of the law of even numbers occurring 
 in the case of the chlorides and bromides of tungsten has 
 been already adverted to (Chap. XVI.). Molybdenum and 
 uranium, like the rest, seem to form a pentachloride. 
 
 The trioxides, MoO 3 , WO 3 , and UO 3 , are generally acid 
 anhydrides. Some of the tungstates of the form M^WC^ 
 seem to be isomorphous with the corresponding chromates 
 and sulphates, but the majority are extremely complex. 
 
256 Chemical Philosophy. 
 
 These trioxides also exhibit a feebly basic function, the 
 oxy-salts of uranium being the most stable. The nitrate, 
 for example, a yellow crystallisable salt, has the formula 
 (UO 2 )"(NO 3 ) 2 .6H 2 O, and the sulphate and oxalate are 
 formed upon the same type. 
 
 Molybdenum and tungsten agree in forming trisulphides, 
 MoS 3 and WS 3 , which are unrepresented among the com- 
 pounds of uranium and of chromium. 
 
 CHAPTER XXV. 
 DIVISION III. METALS. 
 
 METALS are all solids at the temperature of the air, with 
 the sole exception of mercury, which is ordinarily liquid, 
 but solidifies at 40. The solid metals differ very much 
 in their fusibility, the metals of the alkalies melting most 
 readily (potassium at 62'5), whilst platinum and its con- 
 geners require the highest temperature of the oxyhydrogen 
 flame. 
 
 Metals, when in a compact state, combine great opacity 
 with high reflective power, exhibiting the appearance which 
 is sufficiently well known as metallic lustre a property which, 
 however, is exhibited by many other substances, such as 
 iodine, graphite, galena, and many sulphides. Gold in very 
 thin leaves transmits a greenish light. 
 
 That all the common metals are heavy is a fact familiar 
 to every one ; but an inspection of the table of densities 
 given at the commencement of the section will, lead to the 
 conclusion that there are many which are much lighter than 
 the familiar iron, copper, lead, and silver, whilst there are 
 several which are even heavier than gold. As a curious 
 fact, the heaviest solid known is a metal, platinum (reL 
 dens. 21*5), and the lightest known solid is also a metal, 
 namely, lithium, whose relative density is only '578. 
 
Metals. 257 
 
 But' it is of much greater importance to observe that in 
 many cases groups of metals which are associated together 
 by reason of community of chemical properties, have also 
 densities which nearly approximate to one another. 
 
 As examples of this may be cited the alkali metals, 
 lithium, sodium, potassium, rubidium, caesium ; the alkaline 
 earth metals, magnesium, calcium, strontium, barium; the 
 metals of the iron family, chromium, iron, manganese, nickel, 
 cobalt, and copper; silver, lead, and thallium ; and lastly the 
 noble metals, gold, iridium, and platinum. Possibly other 
 similar associations might be discovered, but for the fact 
 that several metals are scarcely known in a compact form, 
 and in other cases the numbers may be more or less in- 
 correct in consequence of the presence of impurity in the 
 specimens operated upon. 
 
 The following are the chief characteristics of metals as 
 a class : 
 
 1. Metals are the best conductors of heat. They differ 
 widely among themselves in this respect, as the following 
 rough comparison shows. Silver, as the best conductor, is 
 placed at the head of the list. The numbers may be taken 
 to represent the relative lengths of bars of equal diameter 
 which, by applying a common source of heat to one ex- 
 tremity, would become equally heated in the same time. 
 
 Silver ...... 1000 
 
 Copper 736 
 
 Gold 532 
 
 Iron . . . . . .119 
 
 Lead 85 
 
 Platinum ..... 84 
 
 Bismuth (an imperfect metal) . 18 
 
 2. Metals are the best conductors of electricity. The 
 metals given in the foregoing list stand in nearly the same 
 order as regards electric conductivity. 
 
 3. Metals are almost always malleable, though in a few 
 
 s 
 
258 Chemical Philosophy. 
 
 cases zinc, for example this happens only at slightly 
 elevated temperatures. Frozen mercury is malleable. 
 
 4. Many metals are ductile, but, since ductility is so 
 largely dependent upon tenacity, it by no means follows 
 that ductility should be manifested by a given metal in the 
 same degree as malleability. Gold, for example, is by far 
 the most malleable metal, though in point of ductility it is 
 surpassed by platinum. Lead and tin, though easily rolled 
 into sheet, can scarcely be obtained in the form of wire, 
 owing to their very slight tenacity. 
 
 5. The oxides of metals are very generally basic (see 
 
 P- 303.) 
 
 The basigenic character belongs exclusively to the metals 
 and metalloids. But the oxygenic function is discharged 
 almost equally well by particular elements in all three 
 divisions. 
 
 The transition from metal to non-metal is, therefore, 
 not accomplished by any sudden break, and the student 
 must be prepared to encounter great, and perhaps insur- 
 mountable difficulties in any attempt to establish a line of 
 demarcation between them. 
 
 The following series of oxides will serve to indicate how 
 gradually the one character disappears as the other is deve- 
 loped. The symbols in black type represent basic oxides, 
 those in roman represent oxides which go to the negative 
 side of the salts into which they enter. 
 
 From Non-metallic to Metallic. 
 
 S0 3 W0 3 Cr0 3 FeO 3 
 S0 2 W0 a 
 
 Cr 2 O 3 Fe 2 O 3 Ni 2 O 3 
 
 No salts. 
 
 CrO FeO NiO CuO 
 
 Extremely 
 oxidisable. 
 
 Cu,O Ag.O K 2 O 
 
 Salts unstable. 
 
Metals of the A Ikalis. 259 
 
 METALS. I. THE METALS OF THE ALKALIS. MONADS. 
 
 Lithium, Li = 7 )Li + K 
 Sodium, Na = 23 I - =23 
 Potassium, K = 39*1) 
 
 Potassium, K =39-1 
 Rubidium, Rb = 85-5 
 Caesium, Cs = 133 
 
 Salt Type, M'CL 
 
 These are soft, white, light, easily fusible, and somewhat 
 volatile metals. They oxidise rapidly in the air, but differ 
 materially from one another in this respect, their affinity for 
 oxygen increasing with the atomic weight. This is shown 
 especially by the spontaneous inflammability of rubidium, 
 and the impossibility of obtaining caesium by decomposition 
 of its carbonate with charcoal. They all decompose cold 
 water with formation of a soluble hydroxide and evolution of 
 hydrogen gas. But as a consequence of the inferior energy 
 of lithium and sodium, the heat developed by their action 
 upon water is not sufficient to cause the ignition of the 
 escaping hydrogen, whereas the hydrogen disengaged by 
 potassium inflames instantly, and continues to burn with a 
 purple light, due to the accompanying vapoui ot the metal. 
 
 The hydroxides of these metals are all fusible, scarcely 
 decomposed by heat, but volatilising at high temperatures, 
 very soluble in water, and the solutions caustic to the skin, 
 alkaline to litmus, and absorbing carbonic acid from the air. 
 The hydroxides also saponify oils and fats, and the resulting 
 soaps are alkaline and soluble. 
 
 The chlorides, sulphides, sulphates, phosphates, and 
 carbonates of the alkali metals are all soluble in water, and 
 their chlorides and sulphates yield perfectly neutral solutions. 
 The sulphates combine with the sulphate of aluminium and 
 the allied metals, generating highly characteristic double 
 
 s 2 
 
260 Chemical Philosophy, 
 
 salts, called 'alums,' which crystallise in octahedrons. A 
 crystallised lithium-alum has not yet been described, but 
 this is probably owing to its great solubility. 
 
 The alkali salts all communicate intense and characteristic 
 colours to the Bunsen flame, and the spectrum (p. 55) of 
 the light so produced exhibits in each case a comparatively 
 small number of bright lines. 
 
 As indicated by the atomic weights, the alkali-metals 
 may be divided into two sub-groups. In the one we have 
 potassium, caesium, and rubidium, which are distinguished 
 by greater chemical activity and inferior solubility of their 
 salts, notably of the platinochlorides and acid tartrates. 
 
 K 2 PtCl 6 KHC 4 H 4 6 
 
 Rb 2 PtCl 6 RbHC 4 H 4 O 6 
 
 Cs 2 PtCl 6 CsHC 4 H 4 6 
 
 Sodium and lithium, though entirely comparable with 
 potassium, are yet distinguished from it by much feebler 
 chemical energies, by the solubility of their platinochlorides, 
 acid tartrates, and alums, and by different minor peculiarities. 
 Sulphate of sodium, for example, crystallises with ten mole- 
 cules of water, whilst sulphate of potassium is anhydrous. 
 The carbonates also differ, that of sodium forming large 
 efflorescent crystals containing ioH 2 O, the carbonate of 
 potassium occurring in small deliquescent granules, contain- 
 ing usually about one or two molecules of water. 
 
 Both sodium and potassium yield unstable peroxides, 
 Na 2 O 2 and K 2 O 2 , and potassium is even capable of passing 
 to a higher stage of oxidation, the product being a tetroxide, 
 K 2 O 4 . No salts corresponding with these oxides exist. 
 
Calcium, Strontium, Barium, 261 
 
 METALS. 2. METALS OF THE ALKALINE EARTHS. DIADS. 
 
 Calcium, Ca = 40 % r B 
 Strontium, Sr = 87-6 ^-= -== 88-5 
 Barium, Ba = 137 ' 
 
 Salt Type, M"Cl* 
 
 These three metals, obtained by the electrolysis of their 
 fused chlorides, are yellowish, hard, and fusible, and ex- 
 tremely oxidisable. They are, however, less oxidisable than 
 the metals of the alkalies, though they are still capable of 
 decomposing cold water. They also communicate charac- 
 teristic colours to flame, and yield spectra (p. 55) which are 
 easily recognisable but somewhat more complex than those 
 of the alkalies generally. 
 
 The hydrates of the alkaline earths are white substances 
 which are decomposed by heat into water and the anhydrous 
 oxides. They are far less soluble in water than the alkalies, 
 and are proportionately less caustic, alkaline, and attractive 
 of carbonic acid. They saponify fats, but the resulting 
 soaps are generally insoluble in water. 
 
 The salts of barium, strontium, and calcium are, as a 
 rule, perfectly neutral. 
 
 The chlorides and sulphides are soluble in water, but the 
 sulphates, phosphates, and carbonates are insoluble. 
 
 Oxide of barium or baryta exposed to a current of air or 
 oxygen at a heat short of redness absorbs oxygen, and be- 
 comes converted into a peroxide, BaO 2 . The corresponding 
 peroxides of strontium and calcium are obtained as white 
 precipitates by adding solution of hydric peroxide to lime 
 or strontia water. All three are resolved by ignition into 
 oxygen gas and a residue of the protoxide. They are also 
 soluble in hydrochloric acid, yielding peroxide of hydrogen 
 and the chloride of the metal, e.g. 
 
 BaO 2 + 2HC1 = BaCl 2 + O 2 H a . 
 
262 . Chemical Philosophy. 
 
 Hence it is presumable that they have the same constitu- 
 tion as hydric peroxide, the oxygen atoms partly satisfying 
 each other's attractions, according to the following graphic 
 
 formulae: Hydric peroxide. Baric peroxide. 
 
 H O /O 
 
 I Ba"/ 1 
 
 H O X O 
 
 Of the three elements, barium, with the highest atomic 
 weight, is decidedly most basylous. Strontium standing next 
 is, in respect to some characters, more nearly related to barium 
 than to the third member of the series, calcium. Thus the 
 nitrates of strontium and barium crystallise in anhydrous 
 octahedrons, isomorphous with lead nitrate, Pb(NO 3 ) 2 . The 
 sulphates of these two metals are also anhydrous, and are 
 practically insoluble in water and acids. 
 
 Nitrate of calcium is a deliquescent salt, soluble in 
 alcohol and crystallising in prisms, which contain 
 Ca(NO 3 ) 2 .4H 2 O. Its sulphate, in the form of gypsum, 
 combines with two molecules of water, is perceptibly soluble 
 in water, and much more freely so in hydrochloric acid. 
 
 On the other hand, strontium agrees with calcium in the 
 production of a deliquescent chloride, which is soluble in 
 alcohol, whilst the barium chloride is insoluble in alcohol. 
 
 The crystallised chlorides have the following formulae : 
 Chloride of Calcium, CaCl 2 . 6H 2 O 
 Strontium, SrCl 2 . 6H 2 O 
 
 Barium, BaCl 2 . 2H 2 O 
 
 METALS. 3. ZINC GROUP. DIADS. 
 
 Magnesium, Mg = 24] M + cd 
 
 Zinc, Zn = 65^^ =68 
 
 Cadmium, Cd = 112] 
 
 Salt Type, M"Cl* 
 
 These elements are rightly associated together in conse- 
 quence of a very obvious sedation of properties, notwith- 
 
The Magnesium- Zinc Group. 263 
 
 standing that they are far less intimately related to one 
 another than, for example, the metals of the alkaline-earth 
 family. Magnesium is a white metal, zinc and cadmium faintly 
 bluish white, and all three are volatile. Their volatility 
 somewhat strangely increases in proportion as the atomic 
 weight increases, whilst their basigenic power diminishes. 
 That zinc is decidedly more positive than cadmium is 
 shown by its power of precipitating cadmium in the metallic 
 state from its solutions. That magnesium is more positive 
 than the other two is shown by its power of decomposing 
 water when heated with it, or more readily if previously 
 coated with pulverulent copper. 1 Also by the precipitation 
 of both zinc and cadmium when metallic magnesium is 
 introduced into solutions of their salts. 
 
 Magnesium, zinc, and cadmium are all easily combustible 
 in air or oxygen, the combustion of the former two being 
 attended by the emission of a dazzling light. Each metal 
 forms one oxide : MgO, a white unalterable powder ; ZnO, 
 a white powder, becoming yellow when heated ; CdO, a 
 yellowish brown powder. These oxides are insoluble in 
 water, that of magnesium only showing a faint alkaline 
 reaction when placed upon wet test-paper. The oxides and 
 hydroxides are readily soluble in solutions of ammoniacal 
 salts. 
 
 The hydroxides are easily resolved, by being heated, into 
 water and the oxide, and the carbonates in like manner give 
 up carbonic anhydride, leaving a residue of the oxide. 
 
 The chlorides are volatile, deliquescent solids. 
 
 The sulphide of magnesium is an earthy substance 
 which is decomposed even by water 
 
 MgS + 2OH 2 = Mg(OH) 2 + SH 2 
 
 and is therefore not precipitated on the addition of a soluble 
 sulphide to a magnesian solution. 
 
 1 Zinc and cadmium, when coated with spongy copper, are also 
 capable of decomposing water slowly. 
 
264 Chemical Philosophy. 
 
 Zinc sulphide (native = blende) is a white precipitate 
 easily soluble in diluted mineral acids, and hence only im- 
 perfectly precipitated by the action of hydrogen sulphide 
 upon the solution of a zinc salt. 
 
 Cadmium sulphide (native = greenockite) is a yellow 
 precipitate thrown down by hydrogen sulphide from acidified 
 solutions of cadmium salts. 
 
 The sulphates of these metals are, perhaps, the most 
 characteristic of their salts. They are soluble in water, and 
 have the following formulae : 
 
 Magnesium sulphate MgSO 4 .7H 2 O 
 Zinc sulphate . . ZnSO 4 .7H 2 O 
 Cadmium sulphate . CdSO 4 .4H 2 O 
 
 The magnesium and zinc salts crystallise in four-sided 
 prisms, isomorphous with the corresponding nickel sulphate. 
 
 All three combine with potassium sulphate, generating 
 double salts, which crystallise with six molecules of water : 
 
 MgSO 4 . K 2 SO 4 . 6H 2 O 
 ZnSO 4 . K 2 SO 4 . 6H 2 O 
 CdSO 4 . K 2 SO 4 . 6H 2 O 
 
 Sulphates isomorphous with these are produced by several 
 other metals, such as copper and iron, and will be referred 
 to in the proper place. 
 
 A general review of its properties indicates that mag- 
 nesium forms a connecting link between zinc and calcium. 
 From the latter it differs in the insolubility of its hy- 
 droxide in water, and in the solubility of the same com- 
 pound in ammonium chloride. Sulphate of magnesium is 
 also distinguished from sulphate of calcium by its ready 
 solubility in water. 
 
Silver and Mercury. 26$ 
 
 METALS. 4. 
 
 Silver, Ag = 108. Mercury, Hg = 200. 
 
 Salt type, M'Cl, Salt types, M'\Cl^ 
 
 orM\Cl^ andM"Cll. 
 
 Silver is a white metal, very malleable and ductile, 
 fusible at a red heat, and volatilising in blue vapour at very 
 high temperatures. It is the best known conductor of heat 
 and electricity. It is undoubtedly related to the metals of 
 the alkalies, for its chloride is isomorphous with sodium 
 chloride, its sulphate is isomorphous with anhydrous sodium 
 sulphate, and by combination with aluminium sulphate it 
 yields a true alum, crystallising in octahedrons. In almost 
 every other respect, however, it differs from them. 
 
 Thus its relative density is high, 10*5, and it is quite un- 
 tarnished by pure air, whether dry or moist, the blackening 
 so often observed being due to the formation of a film of 
 sulphide. Pure silver has the singular property of occluding, 
 when fused in the air, a considerable volume of oxygen, 
 which escapes as the temperature goes down and the metal 
 solidifies. Silver has no action upon water or steam at any 
 temperature, and is quickly reduced from its salts by zinc, 
 iron, or even mercury. The oxide is said to be sufficiently 
 soluble in water to give a faintly alkaline reaction, but it 
 differs entirely from potash or soda in appearance and in its 
 decomposability by heat. The chloride, iodide, sulphide, 
 carbonate, and phosphate of silver are all insoluble in water, 
 and the sulphate very sparingly soluble. 
 
 Silver forms a peroxide, Ag 2 O 2 , even less stable than the 
 potassium peroxide. There is also a suboxide, Ag 4 O, the 
 composition of which is, however, not very well established. 
 
 The relations of silver to the alkali-metals on the one 
 hand, and to mercury on the other, are about equally 
 distant. This connection is obscurely indicated by com- 
 
266 Chemical Philosophy. 
 
 parison of their atomic weights, from which it may be seen 
 that silver is intermediate between sodium and mercury. 
 
 Mercury, like silver, is a white metal, crystallising in the 
 regular system and volatilising, though much more readily. 
 Both are easily reducible to the metallic state, and the 
 metals are both incapable of decomposing water or hydro- 
 chloric acid. Mercury, however, is slowly oxidised by 
 heating in air or oxygen. Both are soluble in nitric acid, 
 and when boiled with concentrated sulphuric acid sulphur 
 dioxide is in each case evolved, and a sulphate produced in 
 the form of a sparingly soluble crystalline powder. 
 
 Silver oxide, Ag 2 O, and mercurous oxide, Hg 2 O, are 
 both nearly black, insoluble in water, and resolved by heat 
 into oxygen gas and the metal. 
 
 Silver chloride, AgCl, and mercurous chloride, Hg 2 Cl 2 , 
 are both obtainable by precipitation as white powders, which 
 blacken, the silver salt quickly, the mercury salt slowly, on 
 exposure to light. 
 
 The iodides are both yellow and insoluble. The chief 
 distinction between silver and mercurous salts lies in the 
 volatility of the latter. 
 
 As regards the formulae of the mercurous compounds, 
 their analogy to the compounds of silver points to the 
 adoption of similar formulae. 
 
 Ag-O-Ag. Hg-O-Hg. 
 
 Ag - Cl, &c. Hg - Cl, &c. 
 
 But, inasmuch as the low vapour density of calomel, 
 117-75 = 20 "i"_.35 5 ? has Deen traced to dissociation 
 
 (Chap. XVIIL), this, and consequently all other mercurous 
 compounds, must be represented as containing diad mercury, 
 
 thus : Hg. Hg Cl 
 
 | >0 i &c. 
 
 Hg/ Hg Cl 
 
Mercuric Compounds. 267 
 
 The relationship subsisting between them and the cor- 
 responding salts of silver is best recalled by using similar 
 formulae for the latter, thus : 
 
 Ag Ag-Cl 
 
 | >0 | &c. 
 
 Ag/ Ag Cl 
 
 Such a view is also consistent with the observed equiva- 
 lency of silver and copper in the sulphides, Ag 2 S, Cu 2 S, and 
 CuAgS. On the other hand, the connection already re- 
 ferred to between silver and monad sodium must not be 
 overlooked. 
 
 The mercuric salts are almost without parallel as regards 
 properties among metallic salts. Their formulas are written 
 upon the same type as those of the salts of cadmium and 
 zinc, but it cannot be said that the resemblance extends 
 much farther. 
 
 In one respect, however, metallic mercury agrees with 
 the metals cadmium and zinc. Notwithstanding its much 
 greater volatility, its vapour density, like that of the other 
 two, is the half of its atomic weight. Hence, whilst the 
 molecules of all other volatile elements are polyatomic, 
 those of the three metals referred to are simply monatcmic, 
 the atom and the molecule being represented in each case 
 by one and the same symbol or formula Zn, Cd, Hg. 
 
 The following are the most important mercuric com- 
 pounds. All are volatilisable. 
 
 Oxide, HgO. A red or yellow powder darkening by 
 heat, and decomposed below redness into Hg and O 2 . 
 
 Chloride, HgCl 2 . White, crystallisable, soluble in 
 water, alcohol, and ether. Vapour density 200 ^ 71 = 135*5. 
 
 Amidochloride, HgNH 2 Cl. Commonly known as 'White 
 Precipitate.' Representable either as an amidochloride, 
 
268 Chemical Philosophy. 
 
 or as chloride of mercuric ammonium, 
 
 Hg 
 
 Cl 
 
 The former view is preferable, considering its formation 
 from mercuric chloride by ammonia. 
 
 /Cl NH 2 H /NH 2 
 
 Hg/ + = Hg/ + 
 
 X C1 NH 3 
 
 Iodide, HgI 2 . Red or yellow, dimorphous, insoluble 
 in water, soluble in ether and in solution of potassic iodide. 
 
 Sulphide, HgS. Cinnabar, dark red masses of hexagonal 
 prisms. Artificial vermilion, bright red powder. 
 
 Sulphate, HgSO 4 . Colourless, minute, prismatic crystals, 
 characteristically decomposed by water with formation of 
 yellow 'turpeth mineral,' or trimercuric orthosulphate, 
 Hg 3 2 S0 4 or H g " 3 (S0 6 ) vi . 
 
 METALS. 5. 
 
 Thallium, Tl = 204. 
 Triad. Salt types M'Cl, and M'"Cl z . 
 
 Lead, Pb = 207. 
 Tetrad. Salt type, M"Cl^ 
 
 THALLIUM was originally discovered and occurs most 
 abundantly in certain kinds of copper and iron pyrites. It 
 is a very soft, nearly white metal, which tarnishes quickly 
 but superficially in the air. It closely resembles lead in 
 physical properties, except that it is softer and streaks paper 
 more readily. It melts at 294, and volatilises at a red heat. 
 It is capable of burning brilliantly in oxygen. 
 
Thallium Compounds. 269 
 
 Thallium forms two classes of compounds, those in 
 which it is univalent being the more stable. 
 
 The following are the most characteristic of thallous 
 compounds : 
 
 Chloride, T1C1. A white curdy precipitate, resembling 
 silver chloride, but crystallisable from boiling water, and 
 combining with platinic chloride to form a yellow, slightly 
 soluble, crystalline compound, Tl 2 PtCl 6 . 
 
 Oxide, T1 2 O, and Hydroxide, T1HO. These compounds 
 are soluble in water, the solution being strongly alkaline, 
 and reacting in the same manner as caustic potash. The 
 hydroxide differs from the alkali hydroxides in losing the 
 elements of water when exposed over oil of vitriol in a 
 vacuum. The residual oxide is almost black, and after fusion 
 crystalline. 
 
 Sulphate, T1 2 SO 4 . A soluble colourless salt, isomorphous 
 with sulphate of potassium. It combines with aluminium 
 sulphate, forming a true alum, TlAl(SO 4 ) 2 .i2H 2 O. 
 
 Sulphide, T1 2 S. A brown precipitate, soluble in acids. 
 
 Thallic Compounds. Chloride, T1C1 3 . A soluble salt, 
 crystallisable by evaporation of its solution in a vacuum. 
 It melts and evolves chlorine at high temperatures. It 
 combines with the monochloride, forming two compounds, 
 T1 4 C1 6 or 3 T1C1.T1C1 3 , and T1 2 C1 4 or T1C1.T1C1,. It 
 forms similar compounds with the chlorides of potassium 
 and ammonium. 
 
 Oxide, T1 2 O 3 . A dark red powder, insoluble in water, 
 and reduced by heat to thallous oxide. Thallium is readily 
 obtained from its salts either by electrolysis, by the action 
 of zinc, or by fusion with cyanide of potassium. 
 
 The ordinary spectrum of thallium is the simplest known. 
 It consists of a single bright band in the green. 
 
 LEAD is a bluish metal, so soft as to streak paper, and 
 possessing very little tenacity. Its specific gravity is 11*36, 
 a number very near to, and intermediate between, that of 
 thallium, 11-9, and of silver, io'53. The latter metal is 
 
270 Chemical Philosophy. 
 
 very constantly associated with it, occurring as sulphide in, 
 probably all, galena. Lead melts at 326, and at a red 
 heat volatilises freely. A freshly cut surface tarnishes rapidly 
 in air, but oxidation, except of the melted metal, does not 
 proceed to any appreciable extent. 
 
 Lead forms two oxides of definite composition, namely, 
 litharge, PbO, and the peroxide, PbO 2 , besides several 
 mixed oxides containing the elements of these two in 
 various proportions. The red oxide is generally 2 PbO, 
 PbO 2 , or Pb 3 O 4 . When plumbic peroxide is acted upon 
 by hydrochloric acid, it does not, like barium peroxide, give 
 rise to hydrogen peroxide, but to a tetrachloride, PbCl 4 . 
 In these two compounds, therefore, lead is quadrivalent, 
 and their formulae must be written as follows : 
 
 CL /Cl 
 O = Pb = O >Pb< 
 
 CK X C1 
 
 The tetrachloride is a very unstable compound, and no 
 oxy-salts corresponding with it are known. 
 
 The ordinary salts of lead are typified by the dichloride, 
 PbCl 2 , a white, sparingly soluble salt, crystallisable from 
 boiling water. 
 
 The oxide, PbO, is a dull yellow, the hydroxide a white 
 powder. Both are slightly soluble in water, the solubility, 
 like that of lime, being increased by the addition of sugar. 
 The solutions are strongly alkaline, and absorb carbonic 
 acid from the air. 
 
 Sulphate, PbSO 4 . In the form of lead-vitriol, isomor- 
 phous with heavy spar, BaSO 4 . 
 
 Carbonate, PbCO 3 . As white-lead ore or cerusite, 
 isomorphous with witherite, BaCO 3 . 
 
 Nitrate, Pb(NO 3 ) 2 , crystallises in octahedrons, isomor- 
 phous with barium nitrate. 
 
 Sulphide, PbS. Black insoluble powder or crystallised 
 in shining cubes (galena). 
 
Thallium and Lead. 2/1 
 
 Relations of Thallium and Lead. 
 
 The relations of thallium both to potassium and lead 
 are exceedingly well-marked. With potassium it agrees in 
 the alkalinity of its oxide and the characters of the platino- 
 chloride, acid tartrate, sulphate, and alum. But it differs 
 from potassium and resembles lead in the ready reducibility 
 of the metal, in its high atomic weight and density, and in 
 the characters of the sulphide, monochloride, and other 
 salts. 
 
 Lead, in its turn, has strong points of resemblance on 
 the one hand to thallium, and on the other to barium, as 
 indicated by the isomorphism of many lead and barium 
 salts. 
 
 The relations which are thus manifested by these metals 
 serve to bring into view the connection which they un- 
 doubtedly have with other metals, the characters of which 
 have already been discussed. Without attempting to trace 
 out any further parallels of the same kind, a task which the 
 student can very well perform without assistance, we may 
 just indicate the direction in which analogies will be most 
 readily detected by arranging the symbols of the most im- 
 portant of these metals in the following order : 
 
 (Hg a )" Hg" Pb" Tl' 
 
 Ag Cd Ba Cs 
 
 | Zn Sr Rb 
 
 Na Mg Ca K 
 
272 Chemical Philosophy. 
 
 METALS. 6. ALUMINIUM GROUP. 
 
 Aluminium, Al = 27 
 Gallium, Ga = 69 
 Indium, In = 114 
 
 Triads or quasi-tetrads. Salt type, MC1 3 or (M v ) vi Cl fi . 
 
 These are white metals of comparatively low density, 
 standing towards one another in a position which is the 
 counterpart of that shown by the group magnesium, zinc, 
 cadmium. 
 
 Aluminium is a very abundant constituent of various 
 silicates. It forms a single oxide and a series of salts which 
 are closely related to the corresponding compounds of iron, 
 as will be shown in the next group. 
 
 Gallium and indium, on the other hand, are very rare 
 elements occurring in certain zinc ores. They differ from 
 aluminium in their lower fusing points and in their more 
 pronounced triad character, also in the formation of more 
 stable sulphides. The following tabular statement will recall 
 their chief characteristics. 
 
 Aluminium Gallium Indium 
 
 Density 2 '6 5-9 7-4 
 
 Melting point 75o-8oo 30 176 
 
 Chloride AlQ 3 (at 760 GaCl, InCl, 
 
 and upwards) 
 
 Oxide A1 2 O 3 white Ga,O 3 white In 2 O 3 yellow 
 
 Alum ^',,}(SO 4 ) 2 I2H.,O Octahedral Octahedral Octahedral 
 
 Sulphide Decomposed Precip. only Yellow pre- 
 
 by water in company cipitate 
 
 with ZnS 
 
Metals related to Iron. 273 
 
 METALS. 7. IRON-COPPER GROUP. 
 
 SUB-GROUP A. 
 Aluminium, Al = 27. 
 Tetrad. Salt type (M^Cl^ 
 
 Chromium, Cr = 52-5. 
 Hexad. Salt types, M" Cl* and (M 2 ) VI CS 6 . 
 
 Manganese, Mn = 55 ) 
 Iron, Fe = 56 J 
 
 Hexad. Salt types, M" Cl^ and (M 2 ) VI C! 6 . 
 
 Cobalt, Co = 587 } 
 Nickel, Ni = 587 J 
 
 Tetrad, Salt type, M" C7 2 . 
 
 These are white or grey metals of very high melting 
 point. Their relative densities follow one another in the 
 same order as the atomic weights, and are almost directly 
 proportional to them. Including copper, the solitary 
 member of the next sub-group, they may be placed in the 
 following order, which we shall show presently represents 
 very fairly their mutual chemical relations : 
 
 Atomic Weights. Relative Densities. 
 
 27 Al 2-6 
 
 52-1 Cr 6-8 to 7-3 
 
 Ate. Wt. Rel. Dens. Ate. Wt. Rel. Dens. 
 
 54 Mn 8-0 (about) 56 Fe 7-8 
 
 59 Co 8-9 58 Ni 8-6 
 
 63*3 Cu 8-9 
 
 T 
 
274 
 
 Chemical Philosophy. 
 
 Al 
 
 Cr 
 
 Fe 
 
 No Aluminous Com- 
 pounds known. 
 
 All Aluminic Com- 
 pounds Colourless. 
 
 CrCl 2 
 CrSO 4 7H,O ? 
 
 pale blue 
 
 FeO 
 FeCl 2 
 FeS0 4 .7H 2 
 
 pale green 
 
 FeSO 4 ) ,-TT n 
 K 2 S0 4 r 6H2 
 
 Cr 3 4 
 
 isomorphous with 
 Fe 3 4 
 
 A1 2 3 
 
 Cr 2 3 
 
 Fe 2 3 
 
 A1 2 C1 6 
 volatile. 
 
 Cr 2 Cl 6 
 
 violet : 
 volatile at red heat. 
 
 Fe 2 Cl 6 
 
 green : 
 volatile at red heat 
 
 Al^SOJ, 
 
 Cr 2 (S0 4 ) 3 
 
 Fe 2 (S0 4 ) 3 
 
 S A1K(SO,) 2 .I2H 2 
 
 CrK(SO 4 ),-i2H,O 
 
 FeK(S0 4 ) 2 .i2H,0 
 
 modified 
 
 
 
 -(FeS 2 ) 
 
 
 Cr0 2 Cl 2 
 
 
 
 
 CrO, 
 
 
 
 
 CrO.K, 
 
 Fe0 4 K 2 
 
 
 Cr0 4 Ba 
 
 FeO 4 Ba 
 
 
 or 
 ,CrO 2 .OK 
 
 I CrO,. OK 
 
 
 
 
 2Cr6 3 2 
 
 
 
 
 or 
 Cr0 2 OK 
 O 
 Cr0 2 
 O 
 CrO 2 OK 
 
 
 
 
 Cr 2 6 (OH),? 
 
 
 
Metals related to Iron. 
 
 275 
 
 Ni 
 
 NiO 
 
 NiCl, 
 
 NiSO 4 .7H,O 
 
 bright green 
 
 Mn 
 
 Co 
 
 MnO 
 
 CoO 
 
 MnCL 
 
 MnS0 4 .7H z O 
 
 pale pink 
 
 Cod, 
 
 CoSO 4 .7HX> 
 
 bright red 
 
 MnSO ) 
 
 CoS0 4l , HO 
 
 K 2 S0 4 V H ^ 
 
 K. 2 SO 4 ) * 
 
 corresponding magnesium salts. 
 Mn,0 4 
 
 No Salts 
 
 Mn,Cl 6 
 
 decomposed by heat 
 
 Mn.(S0 4 ) 3 ? 
 MnK(S0 4 ) 2 i2H,0 
 
 cubes or regular octahedrons. 
 
 -(NiS.) 
 
 MnO 2 
 
 MnO.Cl 2 
 
 Mn0 3 
 
 MnO 4 K 2 
 
 MnO 4 Ba 
 
 Mn 2 O tt (OH) 2 
 and Mn a O 7 
 
 Co 3 4 
 
 decomposed by heat 
 
2/6 Chemical Philosophy. 
 
 The metals of sub-group A resist atmospheric oxidation 
 to a considerable extent, except moisture or carbonic acid 
 be present. Several of them, especially iron and manga- 
 nese, are capable of decomposing steam at a red heat with 
 evolution of hydrogen. They all dissolve more or less 
 rapidly in dilute acids. 
 
 Iron, cobalt, and nickel are strongly paramagnetic, 
 chromium and manganese feebly so. 
 
 A synoptical view of the chemical relations of these 
 metals to one another will be best obtained, since they 
 are somewhat complex, by tabulating the formulae of their 
 most characteristic compounds as on the preceding page. 
 Few observations upon the table are necessary. As in all 
 similar groups of intimately related elements, each one 
 seems to affect a particular state of oxidation or combi- 
 nation in which it attains a condition of chemical repose 
 or equilibrium. Thus the aluminic salts are absolutely 
 irreducible to any lower state of oxidation ; chromic salts 
 reducible with great difficulty ; ferric salts are easily trans- 
 formed into ferrous, whilst manganic salts evolve chlorine 
 or oxygen by mere ebullition. The isomorphous relations 
 of these metals to one another are of great interest, and are 
 exhibited not only by the alums and other artificial salts, 
 but in the constant association of these metals in nature, as, 
 for example, in the replacement of aluminium by iron in 
 clays, and in the silicates from which they are formed, in the 
 replacement of the elements of ferric oxide by chromic 
 oxide in chrome-iron, FeO.Cr 2 O 3 , and by the occurrence of 
 metallic nickel in meteoric iron. 
 
 Several members of the group exhibit peculiarities which 
 are well worthy of study, but in this place can receive only 
 passing notice. Such, for example, are the metameric (?) 
 modifications of alumina and of chromium salts, indicated 
 in the latter case by curious changes of colour and solubility ; 
 also the remarkable influence of a small quantity of carbon 
 in increasing the fusibility and hardness of iron, nickel, 
 
Valency of Iron, Manganese, Chromium, 277 
 
 cobalt, and, perhaps, others of the group ; also the produc- 
 tion of an extensive series of ammoniacal bases by cobalt, 
 not by nickel. 
 
 Valency of the Iron Group. 
 
 Manganese forms a sexfluoride, MnF 6 , and a class of 
 salts, the manganates, which are said to be isomorphous 
 with the chromates and sulphates. Hence manganese is 
 sexivalent, and the formulae of manganic perfluoride and 
 potassic manganate may be written as follows : 
 
 F F KO O 
 
 \ / \ I* 
 
 F Mn-F Mn 
 
 \ 
 
 Manganese also forms a dioxide in which it is quadri- 
 valent, 
 
 O = Mn = O, 
 
 and a chloride corresponding with ferrous chloride in which 
 it may be considered to be quadrivalent, 
 
 C1 2 = Mn = Mn = C1 2 
 
 or bivalent, Cl Mn Cl. 
 
 There is no direct evidence for either the one formula 
 or the other, and, consequently, the latter is very commonly 
 accepted for the sake of simplicity. 
 
 The permanganates must be represented by the formula 
 M' 2 Mn 2 O 8 , on the assumption of the hexad character of 
 manganese, so that potassic permanganate becomes 
 
 O 
 
 \\ 
 O=Mn O OK 
 
 O=Mn O OK 
 
278 . Chemical Philosophy. 
 
 or more probably, 
 
 OK . - OK 
 
 I I 
 
 O=Mn O O Mn=O 
 
 The alleged isomorphism of the permanganates and 
 perchlorates is explainable on the assumption either that 
 chlorine is hexad like manganese and perchloric acid is 
 expressed by the formula H 2 C1 2 O 8 , or that both chlorine 
 and manganese are triad or heptad in the formulae HC1O 4 
 and HMnO 4 , thus 
 
 O O 
 
 / N ii 
 
 H 0-C1 O or H O Cl=0 
 
 \ / II 
 
 O O 
 
 By reason of the correspondence between the ferrates 
 and manganates, iron must also be regarded as hexad ; 
 although in ferric, and probably also ferrous compounds, it 
 is usually assumed to be tetrad. 
 
 Thus the vapour density of ferric chloride at moderate 
 temperatures ! agrees with the half of the molecular weight 
 denoted by the formula Fe 2 Cl 6 , although the formula FeCl 3 
 would equally well record its composition. The existence of 
 iron pyrites, FeS 2 , the analogue of SnS 2 , also supports the 
 hypothesis of the quadrivalence of iron in these compound?. 
 
 Cl Cl 
 
 Cl Fe Fe Cl S=Fe=S 
 
 ! I 
 Cl Cl 
 
 Chromium is also tetrad in its ordinary chromic salts, but 
 
 1 At hi,rh temperatures decomposition takes place into ferrous 
 chloride and chlorine. 
 
Relations ,of Cuprous and Cupric Salts. 279 
 
 hexad in the chromates, as well as in chromium sexfluoride, 
 trioxide, and oxychloride. 
 
 Cl Ci KO O 
 
 II \ / 
 
 Cl Cr Cr Cl Cr 
 
 Cl Cl KO O 
 
 Chromic chloride. Potassium chromate. 
 
 The ordinary salts of cobalt and nickel correspond to 
 the ferrous and zinc salts, but they form unstable peroxides, 
 Co 2 O 3 and Ni. 2 O 3 , which lose oxygen at a red heat. 
 
 These compounds, and the salts of the oxycobaltamines 
 are almost the only known tetradic combinations of these 
 two metals. 
 
 SUB-GROUP B. 
 
 Copper, Ou = 63-3. 
 Diad. Salt-types, (Jf 2 ) ;/ C7 2 and M"Cl* 
 
 This metal is distinguished from the members of the 
 series just considered by its red colour, by its very superior 
 conductivity of heat and electricity, by its diamagnetic pro- 
 perties, by its ready oxidation when heated in dry air, and 
 by the easy reduction of the metal from its oxide and salts, 
 also by its indifference to diluted sulphuric or hydrochloric 
 acid. It is further characterised by the fact that no oxide 
 superior to the oxide CuO can be isolated, by the production 
 of a suboxide, Cu 2 O, and corresponding series of salts, also 
 by the facility with which all its compounds unite with or 
 act upon ammonia. By the last two characters it indicates 
 its relationship to silver and mercury. 
 
 The following are the most important of the compounds 
 of copper, together with the formulae of compounds with 
 which they manifest greater or less analogy : 
 
280 Chemical Philosophy. 
 
 Cuprous. Mercurous* Argentic. 
 
 Cu 2 O Hg 2 O Ag 2 O 
 
 or or 
 
 Cu. Hg v Ag> 
 
 Ag/ 
 Cu Cl Hg Cl Ag Cl 
 
 I I I 
 
 Cu Cl Hg-Cl Ag Cl 
 
 Cu H 
 
 No hydride. No hydride. 
 
 Cu H 
 
 Cu 2 S isomorphous with Ag 2 S and with CuAgS. 
 
 Very few cuprous oxysalts are known, for when cuprous 
 oxide is acted upon by acids it generally splits up and 
 yields a cupric salt and a residue of metallic copper. 
 
 Cupric. Iron and Manganese. 
 
 CuO FeO 
 
 CuSO 4 .5H 2 O isomorphous with FeSO 4 .5H 2 O 
 and with MnSO 4 .5H 2 O 
 
 6H 2 O isomorphous with j^gQ 4 .6H 2 O 
 
 CuS FeS 
 
 FeCuS 2 
 or Fe 2 Cu 2 S 4 FeS 2 
 
 Copper pyrites. Iron pyrites. 
 
 Considering the very general blue or green colour of the 
 cupric salts, and their dissolution in excess of ammonia to 
 form a blue liquid, copper approaches more nearly to nickel 
 than to any other member of the iron group. And although 
 a pentahydrated nickel sulphate corresponding with common 
 blue vitriol has not been described, yet of the isomorphism of 
 copper and nickel in the form of sulphate there can be no 
 doubt. For when a mixture of cupric and nickel sulphates 
 
Platinum Group. 281 
 
 is crystallised, the crystals which are deposited, though con- 
 taining both nickel and copper, have the form of nickel 
 sulphate, and contain seven molecules of water when the 
 nickel is in excess. If, on the other hand, the cupric sul- 
 phate is in excess, the crystals have the form of blue vitriol 
 and contain five molecules of water. 
 
 Through the isomorphism of the double potassio-sul- 
 phates, the iron-copper group of metals is connected with 
 the zinc group previously described. 
 
 METALS. 8. PLATINUM GROUP, 
 
 SUB-GROUP A. GOLD. 
 Triad. Salt types (M 2 )Ct 2 and MCt 3 . 
 
 Gold is a metal of familiar yellow colour and high 
 density. In consideration of its inalterability in the fire or 
 by the action of acids (save nitro-muriatic acid, hence called 
 aqua regia), gold was regarded by the old chemists as the 
 type of a * noble ' metal. It forms two classes of salts, of 
 which the aurous are somewhat unstable, readily undergoing 
 decomposition into an auric compound and metallic gold. 
 
 The sodio-aurous thiosulphate ^ 6 j(S 2 O 3 ) 4 .4H 2 O is one of 
 
 the most definite. 
 
 The trichloride is the most important auric compound. 
 It is a red, crystalline, deliquescent body, which forms yellow 
 crystallisable compounds with the chlorides of hydrogen and 
 the alkali-metals, HAuCl 4 , KAuCl 4 , NH 4 AuCl 4 , &c. 
 
 Gold is reduced from its solutions with the greatest readi- 
 ness by nitrous acid, sulphurous acid, ferrous salts, and all 
 reducing agents, including many organic substances. Its 
 oxides and chlorides also yield up the metal when strongly 
 heated. 
 
282 
 
 Chemical Philosophy. 
 
 SUB-GROUP B. 
 
 The members of this group range themselves very natu- 
 rally into two divisions, each of which exhibits a striking 
 uniformity of atomic weight and density. With one of 
 these gold allies itself, in virtue of its high density and 
 atomic weight, as well as its resistance to oxidation and 
 solution in acids; also, by the yellow colour of its com- 
 pounds and the tendency of its highest chloride to form 
 double salts. With the other, silver is, perhaps, in a similar 
 manner, remotely connected, though in this case it must be 
 admitted there, is but little in the properties of the metal to 
 sanction such an arrangement. 
 
 Symb. At. Wt. Dens. 
 
 Silver Ag - 108 10-5 
 Monad 
 
 Palladium Pd - 106 ii'8 
 
 Rhodium Ro - 104 n-o 
 
 Ruthenium Ru - 104 11-4 
 Tetrad 
 
 Gold 
 Triad 
 
 Symb. At. Wt. Dens. 
 Au - 197 19-3 
 
 Platinum Pt. - 195 21-5 
 
 Iridium Ir - 193 21 'I 
 
 Osmium Os - 199 21-4 
 Tetrad 
 
 The true platinum metals are white metals which require 
 for their fusion the highest attainable temperatures of the 
 oxyhydrogen flame. It is stated, indeed, that osmium has 
 never been really melted ; hence the numbers by which their 
 respective densities are denoted can only be approximately 
 correct. 
 
 They differ somewhat in their behaviour towards re- 
 agents. Palladium, for example, is slowly dissolved by 
 nitric acid, whilst some of the others resist the action even 
 of aqua regia. Ruthenium and osmium, on the other hand, 
 although slowly attacked by aqua regia when in the compact 
 state, undergo oxidation somewhat rapidly when heated in 
 the air. 
 
The Platimim Group. 
 
 283 
 
 They all, with the exception of rhodium, form more or 
 less stable tetrachlorides, which are generally yellow and 
 easily reducible by heat, either to the metallic state or to 
 lower chlorides. These compounds also unite with the 
 chlorides of the alkali-metals, forming crystallisable salts, 
 which in some cases, as, for example, the platinochlorides of 
 potassium and ammonium, 
 
 2 KCl.PtCl 4 or K 2 PtCl 6 
 2NH 4 Cl.PtCl 4 or (NH 4 ) 2 PtCl 6 
 
 are almost insoluble and highly characteristic. But beside 
 these tetrachlorides, the platinum metals yield dichlorides 
 and sesquichlorides, which are in special cases of greater 
 importance. The dichloride of palladium, for example, is 
 one of its most important compounds, and far more stable 
 than the tetrachloride, which is known only in a state of 
 solution, and in its double salts. Rhodium, again, forms 
 only one chloride, and that the sesquichloride. 
 
 The following are all the definite chlorides of these 
 metals which have been described : 
 
 Dichlorides. 
 
 Sesquichlorides. 
 
 Tetrachlorides. 
 
 PdCl, 
 
 
 
 PdQ 4 
 
 PtCL 2 
 
 
 
 PtCl 4 
 
 
 
 Ro,Cl 6 
 
 
 
 IrCL? 
 
 Ir,Cl 6 
 
 IrCl 4 
 
 RuCl, 
 
 Ru C1 8 
 
 RuQ 4 
 
 OsCl, 
 
 Os.Cl, 
 
 OsCl 4 
 
 Ruthenium and osmium are specialised by the existence 
 of potassium salts called rutheniate and osmiate, which 
 contain the elements of potassium oxide, together with a 
 trioxide of ruthenium or osmium. 
 
 K 2 O.RuO 3 or K 2 RuO 4 . 
 
 K 2 O.OsO 3 or K 2 OsO 4 . 
 
 The trioxides, or anhydrides, are not known in an 
 isolated state. But the most remarkable compounds of 
 
284 Chemical Philosophy. 
 
 these metals are the tetroxides RuO 4 , OsO 4 , which are 
 highly volatile bodies, soluble in water to form acid solu- 
 tions, but apparently incapable of generating salts. The 
 constitution of these compounds is not well understood. 
 They are the only tetroxides known. 
 
 CHAPTER XXVI. 
 
 THE PERIODIC LAW. 
 
 THROUGHOUT the preceding chapters, in which the classifi- 
 cation of the elements and of their compounds has been 
 discussed, frequent reference has been made to the remark- 
 able relations which are observable in the numerical values 
 of the atomic weights. These relations have from the first 
 offered to chemists an attractive problem, the importance of 
 which has greatly increased within the last few years. 
 
 Very early in the history of modern chemistry, namely, 
 in the year 1815, the hypothesis was put forward by Dr. 
 Prout that, taking the atomic weight of hydrogen as i, the 
 atomic weights of all the remaining elements are multiples 
 of this by a whole number. This hypothesis, either in its 
 original form or modified subsequently by the assumption 
 of the value *5 or "25 for the atomic weight of hydrogen, 
 has not stood the test of rigorous experimental investigation. 
 But within the last few years a further modification of 
 Prout's hypothesis has attracted attention in connection 
 with speculations relating to the genesis of the so-called 
 elements, based upon the fundamental idea that these bodies 
 were formed in the beginning by the condensation or poly- 
 merisation of the primordial simple matter which is supposed 
 to have pervaded all space. 
 
 Independently of such considerations, however, the 
 relations which have been discovered to subsist between 
 
Atomic Weights of Related Elements. 28 $ 
 
 the atomic weights of the elements and the properties of 
 these substances and of their compounds have been made 
 the basis of a system of classification which has been almost 
 universally adopted by chemists. A short account of some 
 of the most interesting of these observations will now be 
 laid before the student. 
 
 A relation which is conspicuous amongst the atomic 
 weights of the best known elements is one which has already 
 been frequently adverted to in previous pages. It is this. 
 Many closely related elements may be ranged in short 
 series of three, in which a gradual modification of properties 
 is observable in tracing the series through its successive 
 terms. In such series the atomic weight of the inter- 
 mediate term is generally either the half of the sum of the 
 atomic weights of the extremes, or is very near to such 
 number. In the halogens, for example, the atomic weight 
 of bromine is 80, whilst the atomic weight of chlorine is 
 35-5 and that of iodine 127. The mean of these two latter 
 numbers is 81*25. 
 
 A still closer agreement with the mean of the extremes 
 in such a series is shown in the case of the alkali metals, 
 lithium, sodium, and potassium. 
 
 Atomic weight of lithium = 7 
 sodium = 23 
 
 potassium = 39-1 
 
 And 7 + 39-I = 23 . 05 . 
 
 In a few cases the atomic weights of elements which are 
 associated together by reason of their chemical character- 
 istics are simple multiples one of another, as in the example 
 of oxygen, whose atomic weight is approximately 16, whilst 
 that of sulphur is 32, or they have nearly the same value, as 
 in the case of iron and manganese, nickel and cobalt 
 
 These considerations have been very ably discussed by 
 several chemists, but especially by Dumas, who has shown 
 that one of the most interesting relations among the atomic 
 
286 Chemical Philosophy. 
 
 weights of the elements may be expressed somewhat as 
 follows. The numbers representing the atomic weights of 
 a series of closely allied elements, when written down in the 
 order of their numerical value, represent an arithmetical 
 progression which may be expressed by a common formula, 
 
 in which a is the first term and d the difference in passing 
 from term to term. Thus, for example, the series of alkali 
 metals conforms to this type : 
 
 # = 7 ^= 16 
 
 Atomic weight of lithium = a = 7 
 
 sodium = a + d = 23 
 
 potassium = a + id = 39 
 
 Now it is obvious that this relationship is just that which 
 obtains in the combining values of the several members of 
 a homologous series. The radicles of the series to which 
 common alcohol belongs, for example, run in the following 
 order, where a = 15 and d = 14 : 
 
 Methyl CH 3 = 15 = a 
 Ethyl C 2 H 5 = 29 = a + d 
 Propyl C 3 H 7 = 43 = a + id 
 Butyl C 4 H 9 = 57=0 + 3 </ 
 Amyl C 5 H n =71 = a + ^d 
 &c. &c. &c. &c. 
 
 Such a series also exhibits several of the peculiarities 
 already pointed out in several families of elements. Thus, 
 taking three contiguous terms or three terms at equal 
 intervals in the series, the middle one has a combining 
 weight equal to half the sum of the combining weights of the 
 two extreme terms. Thus : 
 
 CH 3 + C 3 H 7 _ 15 + 43 _ 20 _r H 
 - - - 29 . L, 2 tl 5 . 
 
 Moreover in the higher members of such a series numbers 
 such as 141 and 281, 127 and 253, 113 and 225 occur, and 
 
Atomic Weights of Related Elements. 287 
 
 it may be argued that if these radicles were elementary or if 
 their composition were unknown, such numbers might fairly 
 be considered as really representing a ratio of i : 2. 
 
 This simple formula, however, does not account for 
 more than a few of the elementary series, and in many cases 
 it is necessary to recognise a somewhat more complex 
 expression, which may be written 
 
 a + nd + nd' -f nd" + . . . . 
 
 in which a represents the fundamental characteristic of the 
 series, and d, d f ) </", d'" 9 &c., the differences. 
 
 An example or two will render its application easily 
 intelligible. 
 
 THE HALOGENS. 
 
 General Formula. 
 First term a 
 Second a + d 
 Third a + 2d + d' 
 Fourth a + id -f zd' -f d' 
 
 NUMERICAL VALUES. 
 
 First term 19 = F 
 
 Second 19+16-5 = 35-5=0 
 
 Third 19 + 33 + 28 = So = Br 
 
 Fourth 19 + 33 + 56+ 19 = 127 =1 
 
 Here d" = a, and the fourth term therefore becomes 
 
 20. + zd + zd'. 
 
 A similar general formula will be found to apply to the 
 series N, P, As, Sb, Bi. Here a = 14, d = 17, d 1 = 27, 
 d" = 1 8, d'" = 2a. 
 
 Nitrogen = a =14 
 
 Phosphorus = a -f d = 3 1 
 
 Arsenic = a + 2d + d 1 75 
 
 Antimony = a + zd + zd' + d" =120 
 
 Bismuth = a -f 2d + 2d' + ^d" + d!" 
 
 or 30 + 2d + 2d' + 4d" = 208 
 
288 
 
 Chemical Philosophy. 
 
 In the year 1866 a very curious observation was made 
 by Mr. Newlands, to the effect that when the elements are 
 arranged in a continuous series in the order of their atomic 
 weights, commencing with hydrogen, there is at equal inter- 
 vals in the series a recurrence of the same or similar general 
 characters, both physical and chemical. This periodic 
 revival of characteristics occurs, with a few exceptions, at 
 about every eighth member of the series, as will presently 
 be shown. This discovery has been elaborately studied by 
 Mendelejeff and Lothar Meyer, and that which has long 
 been vaguely recognised is now fully established, namely, 
 that the properties of the elements stand in a definite rela- 
 tion to their atomic weights. Further, if we write down a 
 few of the best known elements in the order indicated, and 
 add the formulae of some of their most characteristic com- 
 pounds, the reader who has carefully studied their properties 
 will recognise at once the existence of some kind of periodic 
 relation. 
 
 Element 
 
 Atomic ! Relative 
 weight density 
 
 Atomic 
 volume 1 
 
 Chief 
 chloride 
 
 Chief 
 hydride 
 
 Chief 
 oxide 
 
 Li . 
 
 7 
 
 578 
 
 12 
 
 LiCl 
 
 9 
 
 Li.,O 
 
 Be . 
 
 9 
 
 1-64 
 
 5-6 
 
 BeCl 2 
 
 ? 
 
 Be'O 
 
 B 
 
 ii 
 
 2-68 
 
 
 BC1 3 
 
 BH 3 
 
 B,0 3 
 
 C (Graphite) 
 
 12 
 
 2'2 
 
 5'4 
 
 CC1 4 
 
 CH 4 
 
 CO, 
 
 N . 
 
 14 
 
 
 
 ? 
 
 NH 3 
 
 N 2 O 5 
 
 O . 
 
 16 
 
 Liq. -99 
 
 16 
 
 OC1 2 
 
 OH a 
 
 
 F . 
 
 19 
 
 
 
 
 FH 
 
 
 Na . 
 
 23 
 
 '97 
 
 23 
 
 NaCl 
 
 ? 
 
 Na,O 
 
 Mg . 
 
 24 
 
 17 
 
 14 
 
 MgCl 2 
 
 } 
 
 MgO 
 
 Al . 
 
 27 
 
 2-6 
 
 10 
 
 A1C1, 
 
 ? 
 
 A1A 
 
 Si . 
 
 28 
 
 2-49 
 
 ii 
 
 SiCl 4 
 
 SiH 4 
 
 SiO, 
 
 
 
 
 
 PCL.& 
 
 1 
 
 
 P 
 
 31 
 
 2'I 
 
 
 PC1 5 
 
 j-PH 3 
 
 PS^S 
 
 S . 
 
 32 
 
 2-07 
 
 *{ 
 
 SC1 2 & 
 SC1 4 
 
 }SH 2 
 
 S0 3 
 
 Cl . 
 
 35'5 
 
 Liq. 1-38 
 
 25 
 
 
 C1H 
 
 C1O 2 
 
 K . 
 
 
 865 
 
 45 
 
 KC1 
 
 ? 
 
 K 2 
 
 Ca . 
 
 40 
 
 1-8 
 
 22 
 
 CaCl 2 
 
 9 
 
 CaO 
 
 1 The volume of an atom of an uncombined element in a liquid 
 or solid state is assumed to be in direct proportion to the atomic 
 
The Periodic Law. 289 
 
 The gradual waxing and waning of density and atomic 
 volume, of valency, of disposition to combine with chlorine 
 and with hydrogen, and the basylous or acidulous character 
 of the oxides of these elements may be displayed more 
 conspicuously by a graphic construction such as the fol- 
 lowing. 
 
 Here the curve is traced by the intersection of horizontal 
 lines, whose length is proportional to the atomic weights, 
 and vertical lines, which indicate the atomic volumes. It 
 is certainly remarkable that the atomic volume should 
 not increase with the atomic weight in the same or nearly 
 the same ratio, but should thus increase and diminish in 
 value at pretty regular intervals, but the regularity with 
 which these alternations are followed by the various physical 
 and chemical characters of the elements is still more inte- 
 resting. The student will recognise without difficulty such 
 facts as the following by reference to the diagram. 
 
 It will be seen first that, passing from left to right, the 
 ascending parts of the curve are occupied by elements, the 
 electro-negative character of which becomes more aftd more 
 pronounced the higher they stand, whilst on each summit 
 and the succeeding declivity are stationed elements of 
 electro-positive character which becomes less marked as we 
 go downwards. The elements which are to be found upon 
 the elevated portions of the curve, and which therefore 
 have the largest atomic volume, habitually behave more 
 energetically as chemical agents than those which occupy 
 the lower portions. And this general inference is supported 
 by reference to the heats of formation of the chlorides or 
 oxides of the elements in question. On the ascending 
 portions of the curve, including the topmost points, are also 
 found the volatile and fusible elements, whilst those which 
 occur at the lowest parts are substances which are, for the 
 
 weight, and inversely as the density of the substance : hence atomic 
 volume is an abbreviated expression meaning the quotient obtained by 
 dividing the atomic weight by the density. 
 
 U 
 
290 
 
 Chemical Philosophy. 
 
Mendelejeff's Classification. 291 
 
 most part, fusible and vaporisable only with difficulty. 
 Malleability and ductility and other physical properties are 
 also exhibited periodically at successive points upon the 
 curve. Moreover, it has been found that the properties of 
 the corresponding compounds of the elements, so far as 
 they can be at present traced, exhibit a similar periodic 
 character. Thus the melting points of the normal chlorides 
 of the elements, when set out graphically in the manner 
 shown in the case of the atomic volumes, give a curve 
 which exhibits the same general characters, consisting of 
 alternate elevations and depressions which correspond 
 roughly with those shown in the diagram. The correspond- 
 ing bromides and iodides also give similar curves, following 
 pretty closely the curve of the chlorides. 
 
 The only character of importance which does not exhibit 
 alternate maxima and minima is specific heat. This, how- 
 ever, though not periodic with increase of atomic weight, is 
 somewhat dependent upon atomic volume, for it will be 
 noticed by reference to the diagram that those elements 
 which do not closely conform to the law of Dulong and 
 Petit have not only a small atomic weight, but small atomic 
 volume. 
 
 The table given on p. 292 contains the elements ar- 
 ranged in a manner nearly identical with that proposed by 
 Mendelejeff, and represents a classification based upon the 
 periodic law. It will be noticed that the vertical columns 
 contain closely related elements in two parallel series, the 
 members of which alternate in the value of their atomic 
 weights. The first six horizontal lines are now nearly com- 
 plete, but there are many gaps lower down. It is interesting 
 to notice that the metals gallium, discovered in 1875, and 
 germanium, in 1886, have been found to fill two places pre- 
 viously vacant, and that they present nearly all the charac- 
 ters belonging to the position which, in virtue of their atomic 
 weights, they hold in the system. 
 
 The true places of the rare metals yttrium, lanthanum, 
 
 U 2 
 
2 9 2 
 
 Chemical Philosophy. 
 
 Valency various 
 
 1 1 
 
 u <: 
 
 OSQQ vo if) 
 {J> ^ T3 
 
 Monads or 
 Triads 
 
 io O r> 
 
 2 s r N s- <^. 
 
 Diads or 
 Hexads 
 
 in Os *" f^ ^ 
 
 ^ s I a 
 
 O co co H 
 
 It 
 
 *3" TJ n 
 
 OS o< CO <^. 
 
 S 1 5 3 
 
 $> f-\ O ^_, 00 
 2 P? < CO ^ 
 
 1 
 1 
 
 00 VO < ^- 
 
 y - -X ' e-. co 
 H <fi ^ 
 r^ N ^ o" H 
 
 2 N ^r P! . | 
 
 U co O co cu 
 
 Triads or 
 quasi-triads 
 
 rt- O NH Os ir>00 ^ 
 Os Tj- ro -rf TJ- 'O j^. 
 
 CO <U rt "+ t^ .'-! 
 
 Uh-J fiH w 
 < O H 
 
 tn 
 
 5 
 
 3 $ M ^ g 
 
 O S N U S 
 
 1 
 
 SN S ? - 
 CO 00 01 
 
 co W ^ 
 
 
 C^co^l - iOvOt>' OO OsO*- 1 
 
 hH t < 
 
Exercises on Section V. 293 
 
 cerium, didymium, terbium and erbium are still unsettled, as 
 well as those of ytterbium, decipium, philippium, samarium, 
 &c., the discovery of which has more recently been 
 announced. 
 
 This table requires a few remarks. In the first place 
 there are some elements, as, for instance, iron, manganese, 
 cobalt, and nickel, copper, silver, and gold, for which a place 
 cannot readily be found. Manganese is sometimes con- 
 sidered to stand in the same relation to chlorine that 
 chromium does to sulphur. Silver is undoubtedly allied, 
 though not very closely, with sodium, whilst it is also con- 
 nected with copper on the one hand and with mercury on 
 the other. Gold, again, is unquestionably triad, whilst the 
 platinum metals to which it is most nearly related exhibit 
 even valency. It is also obvious that such relations as 
 are known to exist between thallium and the alkali group, 
 silver and lead, and the special peculiarities of a number of 
 elements, such as mercury, gallium, carbon, and copper, are 
 neither accounted for nor indicated by their position. So 
 that whilst the scheme proposed by Mendelejeff displays 
 the fundamental principle of periodicity very clearly, there 
 are many difficulties in the way of its adoption as a final 
 expression of the true relations subsisting among the ele- 
 ments. 
 
 EXERCISES ON SECTION V. 
 
 1. What experiments and reasoning would lead you to regard 
 chlorine as a monad element ? 
 
 2. In what respects does iodine resemble and differ from bromine ? 
 
 3. Discuss the title of hydrogen to be considered (a) a non-metal ; 
 (b) a metalloid. 
 
 4. Classify the solid elements according to their relative densities. 
 
 5. What are the properties which distinguish chlorine, and the ele- 
 ments most like chlorine, from potassium, and the elements which are 
 most like potassium I 
 
294 Chemical Philosophy. 
 
 6. Compare and contrast the elements phosphorus and arsenic, both 
 in their physical and chemical properties. 
 
 7. What are the special characteristics of carbon which distinguish 
 it from other elements ? 
 
 8. Describe as many as you can of the modified forms of carbon, 
 sulphur, and phosphorus. 
 
 9. What acids are known consisting of hydrogen, oxygen, and phos- 
 phorus ? Give their graphic constitutional formulae, and point out the 
 basicity of each. 
 
 10. What differences and analogies exist between the corresponding 
 compounds of potassium and sodium ? 
 
 11. To what elements does mercury present the greatest resem- 
 blance ? Point out instances of analogy in properties. 
 
 12. Show by a comparison of their compounds in what respects the 
 elements carbon, silicon, tin, and lead may be said to constitute a 
 natural group. 
 
 13. Give the formulae, simple and constitutional, of the following 
 compounds : common salt, caustic potash, sulphuric acid, baryta, 
 barium peroxide, lead peroxide, iron peroxide, silver phosphate, silver 
 arsenate, silver arsenite, kakodyl, and tartar-emetic. 
 
 14. Give a full account of the two oxides of nitrogen known as nitric 
 oxide and nitric peroxide respectively. 
 
 15. Give instances of compounds which do not comply with the law 
 of even numbers. 
 
 1 6. What are the chief points of resemblance and difference between 
 the corresponding calcium and magnesium compounds ? 
 
 17. Assign a place among the elements to each of the following 
 bodies : copper, nitrogen, boron, thallium, vanadium, gold. Give 
 full reasons in each case. 
 
 18. The constitution of phosphorus compounds has been explained 
 on the assumption of the trivalence of the phosphorus atom. Discuss 
 this view. 
 
 19. By what arguments would you support the number 32 against 
 the number 16 as the atomic weight of sulphur? 
 
 20. Caustic potash is often said to be formed upon the water type. 
 Explain this. 
 
 21. What is an alum? At the temperature of 180 common alum 
 loses || of its water. How may this fact be brought to bear upon the 
 construction of its formula ? 
 
 22. Classify some of the better known elements according to their 
 valency, mentioning and explaining, as far as possible, doubtful 
 cases. 
 
Exercises on Section V. 295 
 
 23. Write the formulae of orthosulphuric acid and some of its salts. 
 An ortho-acid may be defined as one in which the characteristic ele- 
 ment (sulphur in this case) is saturated with hydroxyl. 
 
 24. One gram of phosphorus yields 2*2903 grams of phosphoric 
 anhydride. Will this enable you to determine the atomic weight of 
 the element ? and, if not, what further data are necessary ? 
 
 25. How much phosphorus is contained in 120 Ibs. of bone-ash, 
 consisting of Ca 3 (PO 4 ), 88-5 parts and CaCO 3 11-5 parts in 100? 
 
 26. What weight of phosphorus is contained in 10 litres of phos- 
 phorus vapour at 1040? Pressure normal. 
 
 27. What weight of phosphorus is contained in 10 litres of phosphine 
 at 12 ? Pressure normal. 
 
 28. Write the formulae of a few * molecular ' compounds and con- 
 sider in each case whether the compound can be reasonably represented 
 by a single unitary formula. 
 
 29. Calculate the percentage composition of cryolite, Na 3 AlF 6 . 
 
 30. How many kilograms of litharge can be obtained from4 0*5 kilos, 
 of lead, and what volume of oxygen measured at 20 and under 765 mm. 
 is absorbed in the process ? 
 
296 Chemical Philosophy. 
 
 SECTION VI. 
 
 CHAPTER XXVII. 
 
 CLASSIFICATION OF COMPOUNDS. 
 ACIDS, BASES, SALTS. 
 
 Adds. The only element common to all acids is hydro- 
 gen, and the characteristic property of this hydrogen is its 
 ready exchangeability for a metal. It does not appear that 
 this is due to any peculiarity of the hydrogen itself, but 
 to the fact that, in all well-defined acids, it is associated 
 with other elements of very eager affinities. Acids may be 
 divided into two classes according as oxygen does or does 
 not enter into their composition. A classification of acids 
 according to the nature of their constituent elements was 
 given on p. 229. 
 
 It is important to remark, however, that, although such a 
 distinction may for certain purposes be occasionally recog- 
 nised, there is no essential difference between acids which 
 contain oxygen and those which do not. 
 
 Now, it is obvious that the activity with which hydrochloric 
 acid, for example, produces salts when in contact with suit- 
 able metals, is due chiefly to the elective attraction of the 
 chlorine for the metal, and that hydrocyanic acid, a molecule 
 of which contains the same quantity of hydrogen, does not 
 act upon metals with the same promptitude, because the 
 cyanogen group has not the same attraction for metals as 
 chlorine. 
 
 The chemical energy of an acid, then, does not depend 
 
Characters of Acids. 297 
 
 upon the hydrogen which it contains, but upon the electro- 
 negative character, more or less strongly marked, in the 
 elements with which the hydrogen is associated. It is a 
 very remarkable fact, however, that the existence of hydro- 
 gen in a body, in the position which it occupies in acids, 
 namely, in the position of a base, does confer upon the 
 compound one very general characteristic, and that is a sour 
 taste. The sourness of soluble hydrogen salts would, per- 
 haps, have been considered no more remarkable than the 
 astringency of iron salts, the bitterness of magnesium, or the 
 sweetness of beryllium compounds, but that it constitutes a 
 character which serves to differentiate real acids from bodies 
 such as ammonia, alcohols, and metallic hydrates, which all 
 resemble acids so far as to exchange, under certain circum- 
 stances, part of their hydrogen for metals. 
 
 Oxyacids are bodies of very varied character. Some 
 considerations concerning them have already been dwelt 
 upon in the article on Oxygen (Chap. XXIII. , p. 229). 
 
 With regard to the properties of acids in general, suffi- 
 cient has already been written in the article on Hydrogen 
 (Chap. XXIV., p. 245). It now only remains, therefore, to 
 discuss that property of acids known as 'basicity.' This 
 inquiry amounts in some cases to a determination of the 
 weight of the molecule, and reference has already been made 
 to an example of the kind in the article on Molecular Weights 
 (Chap. XXV., p. 131). The question is, however, one of 
 great importance, and deserves to be examined a little more 
 in detail. The basicity of an acid depends upon the num- 
 ber of atoms of hydrogen directly exchangeable out of each 
 molecule for metallic atoms ; an acid containing one such 
 atom of hydrogen being called monobasic, one containing 
 two such atoms is dibasic, or three tribasic, and so on. We 
 may infer the basicity of a given acid from the number of 
 distinct salts it is capable of yielding with each metal. 
 Monobasic acids generally give one salt, dibasic acids, as 
 shown in the case of sulphuric acid, give two, and tribasic 
 
298 Chemical Philosophy. 
 
 acids give three. Mere inspection of the formula of an acid 
 does not, however, furnish sufficient information upon which 
 to form a judgment regarding the basicity of the acid and 
 the number of salts it is capable of producing ; for it by no 
 means follows, from the existence of two or three atoms of 
 hydrogen in the molecule, that it should be dibasic or tribasic. 
 The basicity depends not upon the total number of hydrogen 
 atoms present, but upon the total number endowed with 
 this particular power of metallic exchange. 
 
 The three following acids of phosphorus afford an 
 example of this. Each molecule contains three atoms of 
 hydrogen, but the first is monobasic, the second dibasic, the 
 third tribasic : 
 
 Hypophosphorous acid . H 3 PO 2 or HPH 2 O 2 
 Phosphorous acid . . H 3 PO 3 or H 2 PHO 3 
 Phosphoric acid . . H 3 PO 4 
 
 In this and similar cases it has been generally observed 
 that increase of basicity, as well as of sourness and general 
 chemical acidity, accompanies the addition of oxygen. 
 Hence it has been inferred that those atoms of hydrogen 
 which are exchangeable for metals are combined more 
 intimately with oxygen than others which are not similarly 
 exchangeable. On this hypothesis the formulae of the three 
 acids referred to above are represented in the following 
 manner : 
 
 Hypophosphorous acid . O=P OH 
 
 H OH 
 
 Phosphorous acid . . O=P 
 
 N)H 
 
 x 
 
 Phosphoric acid . . O = P OH 
 
 N)H 
 
Constitution of Sulphuric Acid. 299 
 
 But acids are capable of undergoing a variety of modifi- 
 cations under the influence of reagents, and some of these 
 are available as evidence towards the establishment of their 
 basicity. It has been found, for example, that acids which 
 possess only one basic atom of hydrogen, or are monobasic, 
 yield only one amide and one chloride. Acetic acid, for 
 example, gives the following derivatives, the relation of 
 which to acetic acid is obvious from the formulse 
 
 Acetic acid . . HO.C 2 H 3 O 
 Acetyl chloride . . C1.C 2 H 3 O 
 Acetamide . . NH 2 .C 2 H 3 O. 
 Dibasic acids give two derivatives of the same kind, and 
 tribasic acids give three. 
 
 We may now resume the case of sulphuric acid, adverted 
 to in Chap. XV., as a very instructive example. 
 
 The simplest formula for sulphuric acid is H 2 SO 4 . The 
 problem before us is to prove that this formula, with the 
 relative weight 98, is the formula of the molecule, and that 
 the acid is dibasic. 
 
 i. The fact of the existence of two classes of sulphates 
 has been already appealed to. It is clear that the whole 
 of the hydrogen of the acid is replaceable by metals, and 
 that it is capable of replacement one half at a time. We 
 arrive then at such formulse as these : 
 
 H j 4 Na 
 
 Evidence of this kind is almost conclusive, but it happens 
 that in the present instance other testimony is abundant. 
 
 2. Thus we might show that sulphuric acid is produced 
 when sulphur trioxide and water, both bodies of known 
 molecular weight, unite together 
 
 SO 3 + OH 2 = SO 4 H 2 . 
 
 3. Also when sulphur dioxide is dissolved in water and 
 the solution exposed to contact with oxygen 
 
 SO 2 + OH 2 + O = SO 4 H 2 . 
 
3OO Chemical Philosophy. 
 
 4. It is also formed by the direct union of hydrogen 
 peroxide and sulphur dioxide 
 
 SO 2 + O 2 H 2 = SO 4 H 2 . 
 
 A sulphate being generated in the following strictly 
 parallel case of combination 
 
 SO 2 + PbO 2 = SO 4 Pb. 
 
 5. Sulphuric acid is generated by the decomposition of 
 sulphuryl chloride by water 
 
 S0 2 C1 2 + 2H 2 = S0 4 H 2 + 2HCL 
 
 According to equations 2, 3, 4, the whole of the acting 
 materials enter into the composition of the sulphuric acid 
 which is produced, and therefore there is a strong proba- 
 bility in favour of the molecular weight of sulphuric acid 
 being the sum of their separate molecular weights. Similar 
 considerations apply to equation 5. The molecule of sul- 
 phuric acid is here shown to be built up of the molecules 
 of the sulphur oxychloride and water, minus two atoms of 
 hydrogen and two atoms of chlorine. 
 
 6. We now come to some reactions in which various 
 constituents of the sulphuric acid molecule are replaced. 
 When the acid is heated with phosphoric chloride, two 
 successive reactions of the same kind ensue. In the first 
 an atom of chlorine is introduced in place of the group 
 (OH) which is removed. 
 
 H 2 S0 4 + PC1 5 = HC1S0 3 + POC1 3 + HC1. 
 
 And this kind of exchange is repeated in the second 
 reaction : 
 
 HC1S0 3 + PC1 5 = C1 2 S0 2 + POC1 3 + HC1. 
 
 The relation of the two new chlorides to sulphuric acid 
 may be exhibited most satisfactorily by writing the formulae 
 in the following manner : 
 
 HO^ 2 Cl '^ 2 C1 OW5 
 
 ulphuric acid or Sulphuryl chlor- Sulphuryl 
 
 Hydrate. hydrate. chloride. 
 
Constitution of Sulphuric Acid. 301 
 
 7. Two * amides,' standing in the same kind of relation- 
 ship towards sulphuric acid, also exist. Their formulae may 
 be written thus : 
 
 HO SO HO SQ 
 
 HO bUa NH 2 bU2 NH 2 a 
 
 Sulphuric acid. Sulphamic acid. Sulphamide. 
 
 The existence of these compounds and the correspond- 
 ing chlorides supplies evidence similar to that deducible 
 from a knowledge of the acid and double sulphates (i). 
 These facts tend to show that the hydrogen contained in a 
 molecule of sulphuric acid is divisible into two equal parts, 
 and that these two hydrogen atoms are probably more 
 closely united with two atoms of oxygen than with other 
 constituents of .the molecule. 
 
 8. When concentrated sulphuric acid is made to act on 
 certain hydrocarbons, alcohols, phenols, and other bodies, 
 compounds are produced which have the properties of 
 monobasic acids. These compounds are in some degree 
 analogous to salts ; at any rate, sulphuric acid could not 
 produce derivatives of this kind if it did not contain at 
 least two atoms of basylous hydrogen. 
 
 C 6 H 6 + H 2 S0 4 = H 2 + C 6 H.,(SO 3 H) 
 
 Benzene. Sulphuric acid. Water. Benzene-sulphonic acid. 
 
 C 2 H 5 OH + H 2 SO 4 = H 2 O + C 2 H 5 O(SO 3 H) 
 
 Ethylic alcohol. Sulphuric acid. Water. Ethyl-sulphuric acid. 
 
 C 6 H 5 OH + H 2 SO 4 = H 2 O + C 6 H 4 (OH)(SO 3 H) 
 
 Phenol. Sulphuric acid. Water. Phenol-sulphonic acid. 
 
 All the foregoing facts point to the same conclusion, and it 
 matters not whether we write the formula of sulphuric acid 
 
 H 2 SO 4 or SO 4 or (HO) 2 SO 2 or H 2 O.SO 3 , 
 
 in the attempt to record one or other of its various modes 
 of formation or decomposition. These different expressions 
 represent one and the same molecular weight. 
 
302 Chemical Philosophy. 
 
 In one respect, which has not yet been referred to, there 
 is a difference between acids which contain oxygen and 
 those which do not contain that element. Compounds in 
 which there is no oxygen cannot by any possibility be made 
 to yield water. Hence there are no anhydrides correspond- 
 ing to the haloid acids. Oxyacids, however, may, in various 
 ways, be made to furnish water and an oxide, which is called 
 an anJiydride. Thus sulphurous acid splits up readily in the 
 following manner : 
 
 H 2 SO 3 = H 2 O + SO 2 
 
 Sulphurous acid. Waicr. Sulphurous anhydride. 
 
 Dibasic acids give their anhydrides most readily when 
 simply heated, whereas monobasic acids generally require 
 to be treated with some dehydrating agent. This may, 
 perhaps, be connected in some way with the fact that a 
 dibasic acid always contains within itself the elements of 
 water, whilst a monobasic acid cannot generate water except 
 by the combined action of two molecules. 
 
 Dibasic. 
 H 2 C0 3 H 2 + C0 2 
 
 Carbonic acid. Anhydride. 
 
 H 2 C 4 H 4 4 H 2 + C 4 H 4 3 
 
 Succinic acid. Anhydride. 
 
 Monobasic. 
 
 HNO 3 | TT O a. M O 
 
 HN0 3 } - Ha0 + N2 5 
 
 Nitric acid. Anhydride. 
 
 In the case of some monobasic acids, the anhydride can 
 only be obtained by a succession of operations. In order 
 to prepare acetic anhydride, for example, the chloride is 
 first made from the acid, and this compound is then allowed 
 to react upon a salt. 
 
 C 2 H 3 OC1 + C 2 H 3 O.ONa = NaCl + (C 2 H 3 O) 2 O 
 
 Acetic chloride. Sodium acetate. Sodium Acetic oxide or 
 
 chloride. anhydride. 
 
Definition of l Base! 303 
 
 This difference between monobasic and dibasic acids is, 
 however, not sufficiently general to be maintained as a dis- 
 tinction of much practical importance. 
 
 Bases, The idea implied by the word ' base ' belongs 
 to the obsolete dualistic theory of salts. According to this 
 theory, every salt was supposed to be made up of two parts, 
 one of which, consisting of a metallic oxide, formed the 
 base ; whilst the other part, usually the oxide of a non- 
 metal, or some group consisting of carbon, hydrogen, and 
 oxygen, was regarded as the acid. Sulphate of soda, for 
 example, was, according to dualistic phraseology, composed 
 of the base soda or oxide of sodium, and sulphuric acid or 
 trioxide of sulphur, and its formula was written NaOSO 3 , 
 or, adopting modern atomic weights, Na 2 OSO 3 . 
 
 Now, although this idea has been considerably modified, 
 and we no longer look upon a salt as a double structure of 
 this kind, the fact remains that those oxides of metals which 
 are not overcharged with oxygen do agree in the property 
 of uniting or reacting with acids, so as more or less com- 
 pletely to neutralise them. And this property is shared by 
 the hydrates corresponding with these oxides. Adopting 
 the old name, and taking this neutralising faculty as the 
 criterion of the basic function of oxides, we may define a 
 base as a metallic oxide or hydroxide capable of saturating 
 a. ids. 
 
 It will be noticed that a basic hydroxide and basic oxide 
 have the same relation to each other as an acid and its an- 
 hydride, and consequently it would be quite reasonable to 
 speak of the anhydrous oxide as an anhydride, reserving the 
 term base for the hydroxide. For example 
 
 2 HNO 3 OH 2 = N 2 O 5 
 
 Acid. Water. Acid anhydride. 
 
 2 KHO OH 2 = K 2 O 
 
 Basic hydroxide or base. Basic oxide or basic 
 
 anhydride. 
 
304 Chemical Philosophy. 
 
 H 2 S0 4 OH 2 = S0 3 
 
 Acid. Water. Acid anhyaridft. 
 
 Ca(HO) 2 OH 2 = CaO 
 
 Basic hydrate or base. Water. Basic oxide or basic 
 
 anhydride. 
 
 It must not be forgotten that ammonia and a certain 
 class of its derivatives also possess the power, in a remark- 
 able degree, of uniting with acids and forming perfectly 
 neutral compounds, which present all the external charac- 
 teristics, and many of the chemical properties, of ordinary 
 salts. Ammonia and the alkaloids are therefore commonly 
 regarded as bases, though it is well to remind the student 
 that these bodies differ from metallic oxides or hydroxides in 
 their reactions, both with acids and with anhydrides. Thus 
 with acids, ammonia enters into direct combination, without 
 the formation of any secondary product, thus : 
 
 NH 3 + HNO 3 = NH 3 HNO 3 
 
 or NH 4 NO 3 
 
 Whereas basic oxides or hydroxides invariably generate water 
 in addition to the salt. 1 
 
 KHO + HNO 3 = KNO 3 + H 2 O 
 
 Base. Acid. Salt. Water. 
 
 Again, ammonia combines with acid anhydrides generally, 
 but the compounds which result are not salts, whilst a basic 
 oxide under the same circumstances always yields a salt of 
 the ordinary type. For example : 
 
 BaO + SO 3 = BaSO 4 
 
 Such a reaction is obviously the counterpart of that which 
 occurs when an acid or hydrogen salt is formed from water 
 and the anhydride. 
 
 H 2 + SO 3 H 2 S0 4 
 
 1 The demonstration of this fact is very generally neglected by 
 teachers. An easy experiment by which an appreciable quantity of 
 water may be collected in a few minutes consists in passing dry hydro- 
 chloric acid over gently-heated litharge. 
 
Salts. 305 
 
 The general property of neutralising acids is associated, 
 in the case of all those bases which are soluble in water, 
 with a peculiar soapy taste, and with a power of affecting 
 vegetable colours. The reds are changed to blue or green, 
 the blues to green, and the yellows to brown or brownish 
 red. In litmus the alteration of tint characteristic of alka- 
 line reaction is the result of the conversion of the feeble 
 litmus acid (red) into a blue salt. Somewhat similar changes 
 probably occur in other cases. 
 
 Salts. It has been stated that an acid is a body in 
 which hydrogen is united with a strongly electro-negative 
 element or group of elements. When such a body is acted 
 upon by a metal or metallic oxide, the hydrogen is exchanged 
 for a metal, or something equivalent to it, and a salt is formed. 
 We may define a salt, then, as a compound in which a 
 metallic or other radicle of like nature is combined with 
 a radicle of more or less decided electro-negative character. 
 It has already been explained that electro-negative elements 
 are those which in the electrolysis of the salt are collected 
 at the positive electrode, and which by uniting with hydro- 
 gen form acids. In short, the radicles which, in union with 
 hydrogen, form acids, and with metals form salts, are always 
 composed either of the non-metallic, or, as they are called 
 in this book, oxygenic elements, or of certain metals in 
 company with a relatively large quantity of oxygen. 
 
 A considerable number of salts are known in which a 
 part of the acid radicle of the normal compound is replaced 
 by oxygen or hydroxyl. Such compounds are often called 
 basic salts. The following are some examples which serve 
 to show the general character of these compounds ; 
 
 Plumbic hydroxychloride Pb j 
 Plumbic hydroxynitrate 
 
306 Chemical PJiiljsopJiy. 
 
 (N0 3 
 
 Bismuthic hydroxynitrate BH NO 3 
 
 (OH 
 
 Ferric oxysulphate Fe 2 
 
 Salts may be formed in a variety of ways, amongst which 
 are the following :- 
 
 (a) Union of elements, 
 
 Na 2 + C1 2 = 2NaCL 
 
 (b) Union of acid and basic oxide, 
 
 CaO + CO 2 = CaCO 3 . 
 
 (/:) Action of metal on acid, 
 
 Zn + H 2 SO 4 = ZnSO 4 + H 2 . 
 
 (d) Action of metallic oxide or hydroxide on acid, 
 
 ZnO + H 9 SO 4 = ZnSO 4 + OH 
 NaHO + HN0 3 = NaNO 3 + OH 2 
 
 (e) Double decomposition between two salts, 
 
 BaCl 2 + Na 2 HPO 4 = aNaCl + BaHPO 4 . 
 
 As regards the properties of salts no very general state- 
 ments can be made. Liquid and solid, soluble and in- 
 soluble, crystalline and uncrystallisable bodies are alike in- 
 cluded under this denomination, so that no comparison can 
 now be instituted between the bodies thus named by the 
 chemist and that substance which was probably the proto- 
 type of all salts, namely, common salt (Gr. aA?, Lat. sal, 
 Fr. set, Ger. salz). 
 
 In the production of salts the potential energy of the re- 
 acting materials is more or less exhausted, so that salts are 
 generally bodies of feeble attractions. We observe this in 
 the comparatively small number of compounds into which 
 entire salts enter, and the facility with which such combina- 
 tions are broken up. 
 
Derivatives of Ammonia. 307 
 
 CHAPTER XXVIII. 
 
 DERIVATIVES OF AMMONIA. 
 
 Ammonium Theory. Ammonia is excessively soluble in 
 water, and the solution has all the properties of a powerful 
 alkali. Like caustic potash, it restores the blue colour to 
 litmus which has been reddened by an acid, it absorbs car- 
 bonic acid from the air, it saturates acids, forming a class of 
 crystallisable neutral salts, which are isomorphous with the 
 corresponding salts of potassium. Like potash, also, it pre- 
 cipitates hydroxides when added to the solutions of a great 
 many metallic salts, though in a few cases, e.g. with mercuric 
 salts, it gives rise to compounds of a different character. 
 Lastly, solution of ammonia differs from solution of caustic 
 potash in that when evaporated it leaves no residue. In 
 consequence of this want of stability, solution of ammonia is 
 often referred to as the volatile alkali. 
 
 Guided by this marked resemblance between the com- 
 pounds of ammonia and those of potassium, chemists have 
 been led to regard both series as constituted in the same 
 manner. Writing down in parallel columns a list of ammo- 
 niacal and potash salts, and eliminating from each pair of 
 formulae the symbols which are common to them both, we 
 find that the group composed of one atom of nitrogen and 
 four atoms of hydrogen, NH 4 , is the residue left over in the 
 one case, whilst the solitary symbol K remains in the other. 
 The latter represents the radicle potassium, which, when 
 isolated, presents all the characteristics of a metal. The 
 conclusion seems almost inevitable that the radicle NH 4 
 would also exhibit metallic lustre, conductivity, and so 
 forth, if k could be obtained in a free state. In anticipa- 
 tion of this hypothesis being confirmed, it has been called 
 ammonium. 
 
 X 2 
 
308 ' Chemical Philosophy. 
 
 Compounds of 
 
 Ammonium. 
 
 NH 3 .H 2 O orNH 4 HO or AmHO 
 NH 3 .HC1 or NH 4 C1 or AmCl 
 
 NH 3 .HNO 3 orNH 4 NO 3 or AmNO 3 
 NH 3 .H 2 SO 4 or NH 4 HSO 4 or AmHSO 4 
 (NH 3 ) 2 .H 2 SO 4 or (NH 4 ) 2 SO 4 or Am 2 SO 4 
 NH 3 H 3 PO 4 or (NH 4 )H 2 PO 4 or AmH 2 PO 4 
 (NH 3 ) 2 H 3 PO 4 or (NH 4 ) 2 HPO 4 or Am 2 HPO 4 
 (NH 3 ) 3 H 3 P0 4 or (NH 4 ) 3 P0 4 or Am 3 PO 4 
 
 Potassium. 
 
 KHO 
 
 KC1 
 
 KNO, 
 
 KHSO 4 
 
 KoSO 4 
 
 KH 2 PO 4 
 
 K 2 HPO 4 
 
 K~ 3 P0 4 
 
 A curious phenomenon occurs when a lump of sodium 
 amalgam is placed in a solution of sal ammoniac, or when 
 the same salt, dissolved in water, is submitted to electro- 
 lysis, a globule of mercury being employed as the negative 
 electrode. The mercury swells enormously, becoming so 
 light as to float in the solution, without at the same time 
 losing altogether its metallic aspect. In a few minutes, how- 
 ever, the spongy mass again contracts, hydrogen and am- 
 monia make their escape, and the mercury recovers its 
 original volume and lustre. The body thus produced was 
 long regarded as a true compound of mercury with the 
 hypothetical metallic radicle, and as such received the name 
 ammonium amalgam. It has been shown, however, that 
 the mercury whilst in this condition yields to pressure in 
 the same way as froth, and it is doubtful whether the phe- 
 nomenon is really due to the formation of an amalgam, or 
 to the mechanical distension of the mercury by gas. 
 
 The radicle ammonium has, in fact, never been isolated, 
 and there is no great probability of its ever being obtained. 
 The ammonium theory finds a sufficient basis in the iso- 
 morphism of the ammonia and potash salts, and the general 
 resemblance of the hydrate and its derivatives (to be men- 
 tioned presently) to the fixed alkalis, and the theory itself 
 should be regarded merely as a summary and memorandum 
 of these relations. 
 
 Amines, The hydrogen of ammonia admits of substitu- 
 
Basic Derivatives of Ammonia. 
 
 309 
 
 tion in three successive stages, and the compounds thus 
 formed are believed to be constituted in the same manner 
 as ammonia itself, or, using the ordinary expression, they 
 belong to the ammonia type, 
 
 NJH 
 
 (H 
 
 the replacing radicles being severally combined with the ni- 
 trogen in the same manner as the hydrogen atoms of the type. 
 When positive radicles, consisting of metals or hydro- 
 carbons, are substituted for the hydrogen in the ammonia 
 molecule, the resulting compounds retain the alkalinity and 
 general basic character of ammonia. Such compounds are 
 called amines. They generally combine with acids, but are 
 unaffected by treatment with alkalies. There are several 
 classes of amines resulting from the more or less extensive 
 replacement of the hydrogen or the coalescence of two or 
 more molecules of ammonia. Their constitution will be 
 understood after inspection of the following table, in which 
 the symbol R represents a positive radicle, the valency of 
 which is indicated by dashes : 
 
 Monamines. 
 
 Diamines. 
 
 Triamines. 
 
 Primary. 
 'I K/ 
 
 Nx J H 
 
 LH 
 
 Primary. 
 fR" 
 
 N '{g| 
 
 Primary, 
 j 11"' 
 N 3 H, 
 LH, 
 
 Secondary. 
 ,R' 
 N ! R' 
 
 LH 
 
 Secondary. 
 fR" fR" 
 N ' R" N 2 s J R' 2 
 
 LH, l_H 2 
 
 Secondary, 
 f R'" ( R'" 
 N 3 < R'" N 3 ' R' s , &c. 
 
 LH, LH, 
 
 Tertiary. 
 
 ( R 
 
 N R' 
 
 U' 
 
 Tertiary. 
 fR" fR" fR" 
 N 2 s' R"N J R"NJ R', 
 U" 'U' 2 U' 2 
 
 Tertiary. 
 fR'" i'R"' fR'" 
 N,< R'"N 3 R'"N 3 <; R', 
 
 LR"' IR' S U' 3 
 
 &c. 
 
3IO Chemical Philosophy. 
 
 Monamines are capable of saturating only one molecule 
 of a monobasic acid, but dial-nines unite with two, and 
 triamines with three molecules of such acids. The diamines 
 and triamines also yield intermediate compounds. Repre- 
 senting the monamines, diamines, and triamines respectively 
 by the symbols Am, Am 2 , and Am 3 , the general formulae of 
 the chlorides may be written in the following manner : 
 
 Am.HCl Am 2 .HCl Am 3 .HCl. 
 
 Am 2 . 2 HCl Am 3 . 2 HCl. 
 
 Am 3 . 3 HCl. 
 
 The tertiary amines unite with iodides and other salts 
 of alcohol radicles in the same manner as with acids, fur- 
 nishing compounds which may be formulated either as 
 ammonia compounds or as derived from the hypothetical 
 ammonium radicle. Thus triethyl-amine combines with 
 ethyl iodide, forming a crystalline compound, which may 
 be represented either as N(C a H s ) 3 .C 2 H 5 I, or, more con- 
 sistently and conveniently, as tetrethyl-ammonium iodide 
 N(C 2 H 5 ) 4 L This compound is obviously the analogue of 
 ammonium iodide. When acted upon by silver oxide and 
 water, it exchanges iodine for hydroxyl, and gives rise to a 
 hydroxide of tetrethyl-ammonium, N(C 2 H 5 ) 4 HO, a com- 
 pound which may be obtained in the solid state, and which 
 in its causticity and alkalinity closely resembles potassic 
 hydroxide. The existence of this compound, and others 
 of similar nature, furnishes strong evidence in favour of the 
 ammonium theory. Like their hypothetical prototype, how- 
 ever, the substituted ammonium radicles N(C 2 H 5 ) 4 , &c., 
 are not known in the free state. 
 
 A question has at different times been raised regarding 
 the constitution of ammonium compounds. It seems, how- 
 ever, to be generally admitted that they contain pentad 
 nitrogen, ammonium chloride being represented graphically 
 by the formula- 
 
Constitution of Ammonium Salts. 311 
 
 H H 
 
 \ / 
 
 N 
 
 and other ammonium compounds in a corresponding 
 manner. Experiment has been brought to bear on the 
 subject in the following manner : 
 
 Methyl- triethyl ammonium chloride may be represented 
 either as an ammonium compound 
 
 X 
 
 C 2 H 6 C1CH 3 
 
 or, as formed on the type of sal-ammoniac, regarded as 
 ammonia hydrochloride, NH 3 .HC1. If the latter view is 
 correct, there ought to be two isomeric compounds having 
 the formulae 
 
 N(C 2 H 5 ) 3 .CH 3 C1 and N(C 2 H 5 ) 2 (CH 3 ).C 2 H 5 C1. 
 
 It is found, however, that the result of bringing together 
 triethylamine and methyl chloride, is a body identical in 
 every respect with the product obtained by combining 
 methyl-diethylamine with ethyl chloride. 
 
 According to the ammonium formula, which these 
 results tend to support, each radicle is in the same posi- 
 tion with respect to the nitrogen, and no isomerism is 
 possible. 
 
 Concerning the formation of sal-ammoniac, however, it 
 has been justly remarked, that if we adopt the formula 
 N v (H 4 Cr)., it is difficult to explain why, in the combination 
 
312 Chemical Philosophy. 
 
 of hydrochloric acid and ammonia, the chlorine should 
 be represented as leaving the hydrogen, for which it has 
 great attraction, in order to link itself with nitrogen, for 
 which it appears to have little attraction. 
 
 H H H 
 
 H / \ / 
 
 | + N H = N H 
 
 Cl \ / \ 
 
 II Cl H 
 
 The truth is probably this, that it is the hydrogen of the 
 ammonia, and not the nitrogen, which attracts the chlorine. 
 In other words, it is the entire molecule NH 3 which 
 attracts and combines with the entire molecule HC1. As 
 the result of such combination a new molecule is formed in 
 which the atoms of nitrogen, hydrogen, and chlorine are 
 bound together in a closer and more intimate union than it 
 is possible to express by symbols which assume a limited 
 combining capacity for each element. But this is only one 
 of many cases which show how incompetent are all existing 
 systems of formulae to represent adequately the essential 
 nature of bodies and the chemical changes they undergo. 
 
 Phosphines, Arsines, Stibines. By replacing the hydro- 
 gen of phosphoretted, arsenetted, and antimonetted hydrogen 
 by hydrocarbon groups, extensive series of bases are ob- 
 tained, which have received the above names, and, save 
 for their extreme oxidability, agree in general characters 
 with the compound ammonias. 
 
 Amides. When the hydrogen of ammonia is replaced 
 by oxidised or other negative radicles, compounds called 
 amides are formed. These bodies are, generally speaking, 
 neutral, though some which contain the radicles of weak 
 acids are still capable of entering into combination with 
 acids, and a few are even very decidedly basic ; urea or 
 carbamide, CO (NH.,) 2 for example. 
 
 On the other hand, some of the amides derived from 
 
Amides. 313 
 
 polybasic acids are themselves acid bodies. This has been 
 already referred to (Acids, p. 301). 
 
 The general formulae of amides correspond with those of 
 amines, and they may be classified into primary, secondary, 
 and tertiary monamides,_d_iamides, and triamides, and, sub- 
 stituting negative R or R for R in the formulae given on 
 p. 309, they may be tabulated in the same way as the amines. 
 One slight addition is necessary. Monamines containing 
 bivalent radicles are not known, but monamides in which 
 two atoms of hydrogen are replaced by one bivalent acid 
 radicle, are known under the name irnides. 
 
 Cyanic acid or Carbimide. Succinimide. 
 
 (CO)" N f(C 4 H 4 2 )" 
 
 H w \ H 
 
 Tertiary derivatives of ammonia, called nitriles or 
 cyanides, are known in which the three atoms of hydrogen 
 are replaced by trivalent hydrocarbon groups. We have, for 
 example, 
 
 Formonitrila Acetonitrile Benzonitrile 
 
 N(CH)'" N(C 2 H 3 )'" N(C 7 H 5 )'" 
 
 or Hydrogen cyanide. or Methyl cyanide. or Phenyl cyanide. 
 
 These bodies are not basic, and are decomposed, like all 
 amides, by boiling with alkalis. Ammonia is evolved, and 
 a salt formed corresponding with the acid from the ammo- 
 nium salt of which they are derived by dehydration. Not- 
 withstanding, therefore, that the replacing radicle contains 
 no oxygen, these bodies should be ranked with the amides. 
 The isocyanides (pp. 196 and 238) are basic, and are there- 
 fore called caxbamines. 
 
 The residual hydrogen of primary and secondary amides 
 may be replaced by positive or hydrocarbon radicles. 
 Compounds intermediate between amines and amides then 
 result. The following are examples : 
 
314 
 
 Chemical Philosophy. 
 
 fC 2 H 5 
 
 fC 2 H 5 
 
 fC 2 H 5 
 
 N^H 
 
 N^ C,H 5 
 
 N<^ C,H 5 
 
 IH 
 
 LH 
 
 LC 2 H 5 
 
 Ethylamine. 
 
 Diethylamine. 
 
 Triethylamine. 
 
 rc 2 H 3 o 
 
 rc,H 3 o 
 
 rC 2 H 3 
 
 N<^ H 
 
 N<^ C 2 H 3 
 
 N^ C 2 H 3 O 
 
 LH 
 
 L H . 
 
 I C 2 H 3 
 
 Acetamide. 
 
 Diacetamide. 
 
 Triacetamide. 
 
 rc t H 3 o 
 
 *K C 2 H 5 
 LC 2 H 5 
 
 rc 2 H 3 o 
 
 NfC 2 H 8 
 
 LC.H 5 
 
 rc 2 H 3 o 
 
 N< C 2 H 5 
 
 LH 
 
 Diethylacetamide. 
 
 Ethyldiacetamide. 
 
 Ethylacetamide. 
 
 CHAPTER XXIX. 
 
 CARBON COMPOUNDS. 
 
 BODIES having the same chemical functions, and constitut- 
 ing a series in which, between any two contiguous terms, 
 there is a constant difference of CH 2 , are said to be homo- 
 logous, and the series is a homologous series. Relations of 
 a like nature have not hitherto been observed among the 
 compounds of any element but carbon. 1 This peculiarity 
 seems to be connected with a special faculty with which 
 carbon atoms appear to be endowed, the power, namely, of 
 uniting with one another without the intervention of any 
 other element. Thus a very large number of different com- 
 pounds containing carbon and hydrogen, either alone or 
 associated with oxygen or nitrogen, have already been pro- 
 duced, and a still larger number of possible combinations 
 of carbon is indicated by theory. 
 
 By reason of their composition being unknown or their 
 reactions imperfectly studied, many carbon compounds 
 
 1 And probably silicon. Unless, indeed, we choose to regard such 
 a pair of compounds as tetrachloride and sesquichloride of manganese, 
 MnCl 4 , Mn 2 Cl 6 , as the first two terms of a series Mn^Cl^, + 2 , the higher 
 members of which are unknown. 
 
Hydrocarbons and their Derivatives. 3 1 5 
 
 remain unclassified, but the number of these has been 
 greatly reduced by modern researches. The following are 
 the most important classes of carbon compounds : 
 
 I. HYDROCARBONS AND HALOID DERIVATIVES. 
 
 Some hundreds of compounds consisting only of carbon 
 and hydrogen have been described, but no such compound 
 is known of which two volumes of the vapour contain a 
 larger quantity of hydrogen than is represented by the 
 general formula C M H 2M + 2 . This fact is explained by the 
 following hypothesis. A single atom of carbon has four 
 units of combining capacity, each of which is capable of 
 being saturated by one atom of hydrogen. We thus arrive 
 at the formula H 
 
 H C H 
 
 ! 
 
 H 
 
 which represents marsh-gas. This body is absolutely satu- 
 rated, and nothing will induce it to take up additional ele- 
 ments of any kind, except on condition of removing one or 
 more of its hydrogen atoms. If, now, we imagine another 
 atom of carbon similarly loaded with hydrogen presented to 
 it, the only mode in which they can unite together is by the 
 severance of an atom of hydrogen from each group, as 
 represented in the following diagram : 
 
 H H 
 
 I ! I I 
 
 H-C-JH H-C-H 
 
 I I ' ; I 
 
 H ; H 
 
 or 
 H H 
 
 I I 
 H C C H 
 
 I I 
 H H 
 
316 Chemical Philosophy. 
 
 The same process may be repeated ad infinitum. The 
 number of hydrogen atoms in the first term of the series 
 being 4, in the second 2x4 2, in the third 3x44, in 
 the n th term the number will be 4n (an 2), or 2n + 2. 
 
 The most important series of hydrocarbons at present 
 known are represented by the following general formulae : 
 
 Paraffins, or Marsh Gas Series, C n H. 2n+2 
 
 defines, or Olefiant Gas Series, C n H 2n 
 
 Acetylene Series, C n H 2n _ 2 
 
 Terpene Series, C n H 2n _ 4 
 
 Benzene Series, C n H 2n _ 6 
 
 Naphthalene Series, C n H., n _ 12 
 
 Anthracene Series, C n H. 2n _ 18 . 
 
 From what has already been said as to the manner in 
 which the carbon atoms in a carbon compound become 
 linked together, it is obvious that these general formulae 
 admit of development in a variety of ways, so as to give 
 rise in each series to a considerable number of isomerides. 
 Thus, if we take the formula for the paraffins and assign 
 different values to n we arrive at such results as the 
 following : 
 
 CH 4 n = i 
 
 CH 3 CH 3 n = 2 
 
 CH 3 CH 2 CH 8 ;/ = 3 
 
 CH 3 CH 2 CH 2 CH 3 n = 4 
 
 &c. &c. 
 
 Now it is obvious that in the first, second, and third 
 terms of such a series, on the assumption that H is always 
 monad and C tetrad, there cannot be any other arrangement 
 of the symbols. In other words, there can only be one com- 
 pound of the formula CH 4 , one C 2 H 6 ,and one C 3 H 8 . Iso- 
 merism is impossible. But on arriving at the fourth member 
 of the series, C 4 H 10 , we find it possible to arrange the sym- 
 bols in a second order as follows : 
 
Constitution of Hydrocarbons. 3 1 7 
 
 /CH 3 
 CH 3 CH/ 
 
 The fifth term admits of a still larger number of varieties, 
 which may be expressed in a similar manner. 
 
 i. CH 3 CH 2 CH 2 CH 2 CH 3 
 
 /CH 3 
 
 2. CH 3 CH 2 CH< 
 
 X CH 3 
 /CH 3 
 
 3. CH 3 -C-CH 3 
 
 \CH 3 
 
 As we advance in the series the number of isomeric 
 hydrocarbons theoretically possible increases very rapidly, 
 and it has been calculated that at the tenth term, C 10 H 22 , 
 there may be 75 modifications, whilst at the thirteenth, 
 Ci 3 H 28 , there may be no fewer than 799 varieties of formula. 
 The number of these hydrocarbons actually known is at 
 present very small. 
 
 In the formula for the defines and for the acetylene 
 series we must admit the supposition that the carbon atoms 
 are linked together by more than one unit of valency. 
 
 Ethylene Acetylene 
 
 CH 2 HCH 2 CH=CH 
 
 The terpenes are liquid hydrocarbons found in turpen- 
 tine and in various natural essential oils. They all have the 
 formula C 10 H 1G , and no homologues have yet been dis- 
 covered. Their constitution is at present unknown. 
 
 On reviewing the foregoing formulae it will at once be 
 perceived that whilst the paraffins are saturated, 1 and hence 
 are capable of producing only substitution compounds, the 
 remaining series referred to may enter into direct com- 
 
 1 The paraffins are more completely saturated than any other known 
 form of matter, for not only do they resist the attack of ordinary 
 chemical agents but they form no ' molecular ' combinations, and even as 
 solvents the liquid members of the series have but limited application. 
 
318 Chemical PJdlosophy. 
 
 bination with hydrogen, with the halogens, and other sub- 
 stances. Thus the following formuke represent saturated 
 compounds : 
 
 defines C n H 2n Br 2 
 
 Acetylene series C n H 2n _ 2 Br 4 . 
 
 The formula for benzene may be represented as 
 follows : 
 
 HC CH 
 
 S \ 
 HC CH 
 
 HC=CH 
 
 This formula accounts for the following facts, (i) Ben- 
 zene is the first term of the series, there being no hydro- 
 carbon homologous with benzene containing fewer than six 
 atoms of carbon. (2) Benzene is capable of combining 
 with six atoms of bromine and no more. (3) Benzene 
 gives rise to a large number of substitution derivatives, but 
 of those in which one hydrogen atom is replaced by one 
 monad atom or group there can be no isomerides, because 
 all the six atoms of hydrogen are similarly situated. Of 
 those substitution derivatives in which two atoms of hydro- 
 gen are replaced by two monads there are three isomeric 
 modifications. To explain this we number the points of the 
 hexagon at which the carbon symbols are placed, and 
 omitting for the sake of clearness all but the symbols of 
 the replacing radicles, we see at once in what manner these 
 three modifications of the di-derivatives are related to the 
 original hydrocarbon and to one another. 
 
 Benzene Di-deriv.< lives of Benzene 
 
 i X X X 
 
 X /\ /\ 
 
 I I 
 
 X 
 
 X 
 
Derivatives of Benzene. 3 1 9 
 
 The symbol X represents Cl, Br, I, NO 2 , NH 2 , CH 3 , or 
 other monad radicle. In the first of these formulae the re- 
 placing symbols are represented in the positions i and 2, 
 in the second i and 3, and in the third i and 4. It must, 
 however, be observed that the symbols of carbon to which 
 they are attached are all of the same kind, and it therefore 
 matters not which of these carbons is used as the starting 
 point and in which direction the numbering is carried. The 
 first formula, therefore, merely indicates that the replacing 
 radicles are combined with two contiguous atoms of hydro- 
 gen, whilst the second and third formulae represent them 
 as more widely separated. 
 
 In each case the same structure is implied so long as the 
 relative distances from each other of the symbols of X are 
 preserved upon the chain of carbon. 
 
 The modifications of the tri- and tetra-substitution de- 
 rivatives of benzene are accounted for by the same hypo- 
 thesis. 
 
 The higher homologues of benzene result from the in- 
 troduction of the methyl group, CH 3 , in place of one or 
 more atoms of hydrogen. Monomethyl benzene, commonly 
 called toluene, C 6 Hr,(CH 3 ), cannot exist in more than one 
 form, as already explained. But the next term of the series, 
 xylene, may exhibit no fewer than four isomeric modifica- 
 tions, as follows : 
 
 Ethyl-benzene C 6 H 5 - CH 2 - CH 3 
 Three dimethyl-benzenes C b H 4 (CH 3 ) 2 
 
 Position of methyl 
 
 First modification i 
 
 Second i 
 
 Third i 
 
 The number of modifications is still greater among the 
 higher members of the series. 
 
 The constitution of the hydrocarbons naphthalene, C 10 H S , 
 and anthracene, C 14 H 10 , and their homologues is regarded 
 
320 Chemical Philosophy. 
 
 as closely related to that of benzene, naphthalene being 
 represented as composed of a pair of benzene molecules 
 possessing two atoms of carbon in common : 
 
 CH CH 
 HC f^\ CH 
 
 HC ^ A.J CH 
 
 CH C CH 
 
 In anthracene two molecules of benzene are connected 
 together by a link composed of two additional atoms of 
 carbon, thus : 
 
 CH CH 
 
 HC 
 
 \/c 
 
 CH CH 
 
 CH 
 
 These formulae indicate that the number of isomeric 
 derivatives of naphthalene and of anthracene must be much 
 larger than in the case of benzene. Many of these have 
 already been isolated and their properties determined. 
 They include a number of important dye-stuffs 
 
 Hydrocarbons differ very much in their physical pro- 
 perties, and in the nature of the reactions to which they 
 lend themselves. This, of course, depends upon the series 
 to which they belong ; but taking any one series, such as 
 the paraffins, it is found that a regular gradation may be 
 traced in the physical properties of the several members of 
 the series, passing from the lowest to the highest terms. 
 
 Thus the first three members of the paraffin series are 
 gaseous ; the succeeding members are liquid, but become 
 less and less volatile, 1 and at the same time more dense 
 and viscid, till among the highest members of the series we 
 
 1 See Chapter V., p. 35. 
 
Alcohols. 321 
 
 come to crystalline solid bodies, which are volatilisable 
 only at such high temperatures that they cannot be distilled 
 without extensive decomposition. In proportion as we 
 ascend the series the percentage of hydrogen rapidly de- 
 creases, and this, together with the diminished mobility of 
 the body, may perhaps account for the comparative in- 
 difference to chemical reagents exhibited by the higher 
 members. 
 
 A corresponding seriation of properties, both physical 
 and chemical, is observed in other groups of bodies related 
 to one another in a similar manner. 
 
 2. ALCOHOLS. 
 
 These bodies may be regarded as hydroxides of hydro- 
 carbon radicles, considering that in their most prominent 
 and characteristic reactions they resemble water and the 
 hydroxides of the metals. They are either liquids or solids, 
 and they are obtained in a variety of processes ; but the 
 ethylic or common alcohol of fermentation was the first 
 studied, and is the best known of the class. Taking it as 
 the representative of alcohols in. general, we may record the 
 following as its characteristic reactions : 
 
 (a) By the action of an alkali metal it loses hydrogen. 
 
 C,H 5 HO + K = C 2 H,KO + H. 
 
 Alcohol. Potassic ethylate. 
 
 (b) By the action of acids or acid anhydrides it is con- 
 verted into salts, called compound ethers, e.g. : 
 
 C 2 H 5 HO + HC1 = 
 
 Alcohol. 
 
 C 2 H 2 C1 + 
 
 Ethyl chloride. 
 
 H 2 O. 
 
 C 2 H,HO + H 2 S0 4 = 
 
 Alcohol. 
 
 C 2 H,HSO 4 + 
 
 Ethyl-hydrogen 
 sulphate. 
 
 H 2 0. 
 
 ,HO + (C 2 H 3 0) 2 = 
 
 'hoi. Acetic anhydride. 
 
 2C 2 H 5 C 2 H 3 O 2 
 
 Ethyl acetate. 
 
 + H 2 0. 
 
 
 
 Y 
 
322 Chemical Philosophy. 
 
 (c) The alcohol is reproduced from these salts by the 
 action of alkalies : 
 
 C 2 H 5 C1 4- KHO = C 2 H 5 HO + KC1. 
 
 Alcohols are divisible into several classes, according to 
 the number of hydroxyl groups in the molecule. Mono- 
 hydric alcohols yield one saline derivative or compound 
 ether, dihydric alcohols give two, trihydric alcohols give 
 three such compounds, and so on. 
 
 The following formulae, which represent three typical 
 alcohols and the chlorides derivable from them, will serve 
 to illustrate this statement. 
 
 Common alcohol. Glycol. fUyceroL 
 
 Monohydric. Dihydric. Trihydric. 
 
 (HO HO 
 
 C 2 H 5 HO ^ 2 
 
 (HO 
 C 3 H 6 HO 
 
 C H 1 HO Q 
 
 ^ 2tl4 lCl (HO 
 
 C 8 H 5 Cl 
 Id 
 
 rn 
 
 CTT /-H /- TT I *^1 
 
 ojn = \^,i v^on.4 -) t~*i 
 
 la 
 
 But alcohols are also divisible into several other groups, 
 which are characterised by their behaviour when submitted 
 to oxidation. 
 
 Primary alcohols, when oxidised, lose hydrogen, and 
 yield bodies which are called aldehyds ; thus common alco- 
 hol loses two atoms of hydrogen : 
 
 C 2 H 6 + O = C,H 4 + OH 2 . 
 
 Alcohol. Aldehyd. 
 
 Secondary* alcohols, under the influence of similar re- 
 agents, yield ketones ; for example : 
 
Primary, Secondary ', and Tertiary Alcohols. 323 
 C 3 H 8 + O = C 3 H 6 + OH 2 . 
 
 Isopropylic alcohol. Acetone. 
 
 Tertiary alcohols, when oxidised, break up into a mixture 
 of acids, each containing a smaller number of carbon atoms 
 than the alcohol. 
 
 These differences are accounted for by the following 
 hypothesis. The generating hydrocarbon is supposed to 
 be marsh-gas or methane, and by the replacement of one 
 of its hydrogen atoms by hydroxyl, an alcohol called methy- 
 lic alcohol is formed. 
 
 Methane Methylic hydroxide or methylic alcohol. 
 
 H 
 
 H 
 
 H 
 
 IH 
 
 H 
 H 
 H 
 OH 
 
 By replacing another atom of hydrogen by a hydrocarbon 
 group, one of the higher homologues of methylic alcohol is 
 produced. Now, so long as the replacement is effected 
 in such a manner as to leave two of the hydrogen atoms of 
 the marsh-gas type undisturbed, the product is a primary 
 alcohol, which by the substitution of an atom of oxygen for 
 these two atoms of hydrogen is capable of giving rise to an 
 aldehyd. 
 
 Aldehyds. 
 
 Primary alcohols. 
 
 [CH, 
 
 ,C(CH 3 )H 2 
 
 C H 
 
 H 
 H 
 
 lOH 
 
 IOH 
 
 Ethylic. 
 
 Propylic. 
 
 Acetic. Propionic. 
 
 If, however, the second step in the process of replace- 
 ment affects one of these typical hydrogen atoms, then a 
 secondary alcohol is produced, and this body cannot take 
 in oxygen in the same manner as the primary alcohols, but 
 it may lose two atoms of hydrogen, as represented by the 
 
 following formula 
 
 Y2 
 
324 Chemical Philosophy. 
 
 Secondary alcohol. Ketone. 
 
 CH : 
 
 CH : 
 
 C ] 
 I OH (" 
 
 Propylic. Acetone or dimethyl-ketone. 
 
 Lastly, the tertiary alcohols are generated in the same 
 manner by the replacement of the last atom of hydrogen 
 originally belonging to the type, thus : 
 
 Tertiary alcohols. 
 
 (CH 3 (C(CH 3 )H 2 
 
 /-* v^rio o 
 
 OH VOH 
 
 Butylic. Amylic. 
 
 It is obvious from this that, although a primary alcohol 
 may contain one or two atoms of carbon, no secondary 
 alcohol containing less than three atoms, and no tertiary 
 alcohol containing less than four atoms of carbon in the 
 molecule can exist. 
 
 Another class of alcohols, called phenols, is also known. 
 These agree to some extent with secondary alcohols, inas- 
 much as they yield no aldehyd by oxidation, but furnish 
 compounds containing the same number of carbon atoms as 
 themselves. These substances, however, are not ketones. 
 They are called quinones. 
 
 No phenol is known to contain fewer than six carbon 
 atoms in the molecule. They are regarded as derived from 
 benzene or some analogous hydrocarbon. The best known 
 phenol is the body which usually goes by that name among 
 chemists, or, in common parlance, carbolic acid, C 6 H 5 OH. 
 In accordance with the formula, it might also be called 
 phenyl hydroxide, or phenyl alcohol. It differs from ordi- 
 nary alcohols in yielding a great number of substitution 
 derivatives, most of which exhibit well-marked acid pro- 
 perties. Thus the trinitrophenol is commonly called picric, 
 or trinitrophenic acid. The following are the formulae of its 
 chloro- and nitro derivatives : 
 
Ethers Aldehyds and Ketones. 325 
 
 Monochlorophenol . C 6 H 4 C1.OH 
 
 Dichlorophenol . C 6 H 3 C1 2 .OH 
 
 Tnchlorophenol . C 6 H 2 C1 3 .OH 
 
 Pentachlorophenol . C 6 C1 5 .OH 
 
 Mononitrophenol . C 6 H 4 (NO 2 ).OH 
 
 Dinitrophenol . C 6 H 3 (NO 2 ),.OH 
 
 Trinitrophenol . C 6 H 2 (NO 2 ) 3 .OH 
 
 Several isomeric modifications of the mono- and di-deriva- 
 tives are known. 
 
 3. ETHERS. 
 
 These bodies have the same relation to alcohols as the 
 metallic oxides to their hydrates. They are, however, not 
 easily converted into alcohols by the direct action of water, 
 neither are they obtained from alcohols so readily as the 
 metallic oxides from hydrates. The oxides corresponding 
 with the dihydric alcohols seem to undergo this transforma- 
 tion more readily than those belonging to the monohydric 
 alcohols. The following are some examples of simple and 
 mixed ethers : 
 
 CH 3 } CH,} C 2 H 
 
 CH 3 j u C 2 Hj u C 2 H 
 
 Methylic oxide or ether. Methyl-ethylic Ethylic oxide (common 
 
 oxide or ether. ether). 
 
 (C 2 H 4 )"0 (C 8 H S ) 2 " 3 
 
 Ethylene oxide or ether. Glyceryl oxide or ether. 
 
 The known ethers are volatile liquids of aromatic odour. 
 
 4. ALDEHYDS AND KETONES. 
 
 These two classes of compounds are closely related, both 
 as regards their mode of formation, and properties and 
 composition. 
 
 Aldehyds are formed by the oxidation of primary alcohols, 
 ketones by the oxidation of secondary alcohols. Thus 
 common aldehyd results from the removal of two atoms of 
 hydrogen from common (ethylic) alcohol : 
 
 C 2 H 6 - H 2 = C 2 H 4 
 
326 
 
 Chemical Philosophy. 
 
 Acetone, the best known of the ketones, stands in the 
 same relation towards isopropylic alcohol : 
 CaHgO H 2 = C 3 H 6 O 
 
 It has already been shown that no secondary alcohol can 
 contain less than three atoms of carbon, and consequently 
 there can be no ketone, properly so called, containing less 
 carbon than acetone, C 3 H 6 O. 
 
 The formation of this body by the action of zinc methyl 
 (see organo -metallic compounds] on carbonyl chloride shows 
 that it contains two methyl groups united with the carbonyl 
 radicle : 
 
 cl + 
 
 co 
 
 Carbonyl chloride. Zinc methyl. 
 
 Zinc chloride. 
 
 Acetone. 
 
 It thus appears that the aldehyds of the monocarbon and 
 dicarbon series may be considered to be homologous with the 
 higher aldehyds on the one hand, and with the ketones on the 
 other. The following formulae indicate the relationship : 
 
 Formic aldehyd ^Q f H 
 
 or ketone. I H 
 
 Aldehyds. 
 
 Propionic. 
 
 Ketones. 
 
 Dimethyl (Acetone). 
 
 CO 
 
 Butyric. 
 
 c 3 H 7 
 
 Valeric. 
 
 Methyl-ethyl. 
 
 >J^| (B.P. 81) 
 
 Diethyl. 
 
 O 
 
 Methyl-propyl. 
 CHq /-r, n 
 
 Methyl-isopropyl. 
 
AldeJiyds and Ketones. 327 
 
 From the existence of such intimate relations between 
 these two classes of compounds it is not extraordinary that 
 'in many of their properties there should be considerable 
 resemblance between them. 
 
 (a) By the action of nascent hydrogen, generated by 
 the action of sodium amalgam in the presence of water, 
 aldehyds and ketones are converted into the corresponding 
 alcohols, e.g. : 
 
 C,H,O + H 2 = C 2 H 6 O 
 
 Acetic aidehyd. Ethyl alcohol. 
 
 C 3 H,.,O + H a = C :1 H H O 
 
 Acetone. Isopropyl alcohol. 
 
 (b) All aldehyds and many ketones combine with the 
 acid sulphite of sodium, forming crystallisable compounds, 
 which yield up the aldehyd or ketone again by treatment 
 with an acid or an alkaline carbonate. 
 
 (f) By boiling with hydrocyanic acid and hydrochloric 
 acid, aldehyds and ketones are converted into acids con- 
 taining in the molecule one more atom of carbon than 
 themselves. For example, common aldehyd is converted 
 into one form of lactic acid : 
 
 CO 
 
 CH 
 H 
 
 3 + HCN + 2 H 2 + HC1 
 
 Aldehyd. Hydrocyanic acid. 
 
 ,QH 
 
 CH 3 
 
 1CH 
 
 + NH 4 C1 
 ^CO.OH 
 
 Ethylidene lactic acid. Ammonium chloride. 
 
 Acetone is converted by a similar process into a homo- 
 logue of lactic acid : 
 
 + HCN + 2 H 2 O + HC1 
 
 Acetone. Hydrocyanic 
 
 acid. 
 
 (CH 3 
 
 - / CH3 
 
 iCfOH + NH 4 C1 
 
 I \CO.OH 
 
 Acetonic or Ammonium 
 
 oxybutyric acid. chioride. 
 
328 C/iemical Philosophy. 
 
 Notwithstanding these various points of resemblance, 
 ketones are sharply distinguished from aldehyds by the 
 action of agents of oxidation. 
 
 An aldehyd is always converted by oxidation into an acid 
 containing the same number of atoms of carbon. 
 
 A ketone, on the contrary, is much more difficult to 
 oxidise, and is then broken up into a mixture of acids, each 
 containing a smaller number of carbon atoms. This oxida- 
 tion proceeds according to a definite law. 
 
 5. ACIDS. 
 
 Every carbon acid may be considered to have been 
 derived from an alcohol by the substitution of an atom of 
 oxygen for two atoms of hydrogen. 
 
 Monohydric alcohols yield monobasic acids. For ex- 
 ample, ethylic alcohol yields acetic acid : 
 
 C 2 H 5 .OH - H 2 + O = C 2 H 3 O.OH 
 
 Ethyl alcohol. Acetic acid. 
 
 Dihydric alcohols may yield either monobasic or dibasic 
 acids. Ethylene alcohol or glycol, for example, may be con- 
 verted into monobasic gly colic or dibasic oxalic acid, thus : 
 
 C 2 H 4 (OH) 2 - H 2 + O = C 2 H 2 0(OH) 2 
 
 Ethylene alcohol. Glycolic acid. 
 
 C 2 H 4 (OH) 2 - 2 H 2 + 2 = C 2 2 (HO) 2 
 
 Ethylene alcohol. Oxalic acid. 
 
 Trihydric alcohols may yield theoretically either mono- 
 basic, dibasic, or tribasic acids. Very few of these deriva- 
 tives, however, are actually known. Glycerol, a trihydric 
 alcohol, yields, upon oxidation, monobasic glyceric acid and 
 dibasic tartronic acid : 
 
 CH 2 .OH CH 2 .OH CO.OH 
 
 H.OH CH.OH CH.OH 
 
 I I I 
 
 CH 2 .OH CO.OH CO.OH 
 
 Glycerol. Glyceric acid. Tartronic acid. 
 
Acids. 329 
 
 A large number of acids are known, the alcohols corre- 
 sponding to which have not yet been discovered. 
 
 Acids, however, may be generated by an entirely different 
 process from alcohols containing one atom less carbon. 
 This process is interesting, because it serves to indicate the 
 general constitution of carbon acids. 
 
 Starting from methyl alcohol, CH 3 .OH, for example, we 
 may convert this body into the corresponding cyanide, 
 CH a .CN. When boiled with caustic potash this compound 
 undergoes decomposition, ammonia is evolved, and the 
 potassium salt of acetic acid is generated : 
 
 CH 3 .CN + KHO + H 2 = CH 3 .CO(OK) + NH 3 
 
 Methyl cyanide. Potash. Water. Potassium acetate. Ammonia. 
 
 Or, leaving the potassium out of sight, the reaction may 
 be represented as follows : 
 
 CH 3 .CN + 2H 2 O = CH 3 .CO(OH) + NH 3 
 
 From this it appears that the acid is generated by the 
 substitution of O" + (OH)' for the N'" of the cyanide. 
 Acetic acid, then, consists of a methyl group, CH 3 , com- 
 bined with a carboxyl group, CO.OH. Now methylic 
 alcohol is monohydric, and therefore is capable of yielding 
 one cyanide and no more. Consequently, from methylic 
 alcohol only one derivative containing the carboxyl group 
 can be formed, and that derivative is a monobasic acid. In 
 a similar manner it is found that, by decomposing the 
 cyanide of a bivalent alcohol radicle by boiling alkali, a 
 bibasic acid is produced, whilst the cyanide of a trivalent 
 radicle yields by the same process a tribasic acid. Setting 
 down examples of compounds generated in this way, we 
 find that they all contain the group CO.OH, once, twice, or 
 three times in the molecule, according to the saturating 
 power of the acid. 
 
330 Chemical Philosophy. 
 
 Monobasic Dibasic Tribasic 
 
 acetic acid. succinic acid. 1 tricarbailyiio acid. 
 
 CH 3 .CO(OH) C 2 H 4 {CO(OH) C 3 H 5 {co|oHJ 
 
 This carboxyl group, then, is contained in a very large 
 number of carbon acids, and its presence seems to be 
 intimately connected with the development of the acid 
 character in these compounds. We have seen (Phenols, 
 p. 324) that the halogens and the nitroxyl group (NO>_,)', when 
 substituted for hydrogen in certain oxidised bodies, also con- 
 stitute acidifying agents, but the presence of these radicles in 
 varying quantities does not affect the basicity of the result- 
 ing acid ; whereas it seems to have been established that 
 an acid which contains the carboxyl, CO(OH), group n 
 times is n basic. 
 
 Carbon acids, when submitted to the action of the halo- 
 gens, yield a great many substitution derivatives, such as the 
 following chlorinated compounds, which are obtained from 
 acetic acid : 
 
 Monochloracetic acid CH 2 C1.COOH 
 Dichloracetic acid CHC1 2 .COOH 
 Trichloracetic acid CC1 3 .COOH 
 
 But here the process of substitution stops, the hydrogen 
 of the carboxyl being replaceable only by metallic or posi- 
 tive radicles, and not by negative bodies. 
 
 The metallic salts, amides, anhydrides, and chlorides 
 derived from a carbon acid, correspond in number, mode of 
 formation, and general properties with the same derivatives 
 of mineral acids. 2 
 
 6. BASIC DERIVATIVES OF AMMONIA. 
 
 The constitution and general properties of these bodies 
 have already been discussed (Amines, p. 309). 
 
 1 See Molecular Weights, Chap. XVI., p. 129. 
 
 2 Pp. 296 to 303. 
 
Basic Derivative* of A mmonia. 331 
 
 The 'alkaloids,' quinine, morphine, strychnine, and many 
 others which are obtained from various plants, and con- 
 stitute important medicinal agents, are referable to this type. 
 They are for the most part tertiary amines. 
 
 7. COMPOUND ETHERS OR ETHEREAL SALTS. 
 
 The most general method for the formation of these com- 
 pounds consists in heating together an acid and an alcohol. 
 A double decomposition ensues ; water is formed, together 
 with the ethereal salt ; so that the reaction is exactly 
 parallel with the change which occurs when an ordinary 
 acid and basic oxide are brought into contact. The fol- 
 lowing equations, for example, are strictly comparable : 
 
 KHO + HNO 3 = KNO 3 + H 2 O 
 
 Basic oxide. Acid. Salt. Water. 
 
 C 2 H S .HO + HC 2 H 3 O 2 = C 2 H 6 .C 2 H 3 O 2 + H 2 O 
 
 Alcohol. Acid. Ethereal salt. Water. 
 
 Compound ethers also imitate metallic salts in their 
 general reactions. Thus they are decomposed by acids, by 
 alkalies, and by water in precisely the same manner. They 
 differ from metallic salts, however, in many physical cha- 
 racters, being for the most part volatile liquids of aromatic 
 odour, and very slightly soluble in water, though generally 
 miscible with alcohol in all proportions. Comparatively 
 few are crystallisable. Among the most interesting examples 
 of crystallisable ethereal salts are the constituents of the 
 natural fats : 
 
 Palmitin C 15 H 31 .CO.O\ 
 
 or glyceiic palmitate C,f,H 31 .CO.O-(C 3 H 5 )''' 
 
 L^arisi.vA'.u/ 
 and 
 
 Stearin or C 17 H 35 .CO.O\ 
 
 stearin or PR rn n ir "H" y 
 
 glyceric stearate rw' rn n / 3 ' 
 
33 2 Chemical Philosophy. 
 
 8. ORGANO-METALLIC COMPOUNDS. 
 
 These are compounds of metals or metalloids with 
 hydrocarbon radicles. They are volatile, very oxidisable, 
 and generally heavy liquids. The zinc-ethyl compound was 
 first obtained. It is a colourless liquid, boiling at 118, and 
 spontaneously inflammable when thrown into the air. It is 
 instantly decomposed by water, and when submitted to the 
 action of successive small quantities of oxygen it yields two 
 oxidised compounds, thus : 
 
 Zinc ethide. Zinc ethyl-ethylate. Zinc ethylate. 
 
 7 n jC 2 H 5 7 |C 2 H S 7 (OC 2 H 5 
 
 ' n (C 2 H 5 ' n (OC 2 H 5 <n lOC 2 H 5 
 
 Iodine removes from it half and then the whole of the 
 ethyl, according to these equations : 
 
 Zn(C 2 H 5 ) 2 + I 2 = Zn(C 2 H,)I + C 2 H 5 I 
 
 Zinc ethiodide. Ethyl iodide. 
 
 Zn(C 2 H 5 )I + I 2 = ZnI 2 + C 3 H 5 I 
 
 Zinc iodide. Ethyl iodide. 
 
 The existence of these compounds furnishes conclusive 
 testimony of the diadic character of the zinc atom, and in 
 other cases similar evidence, of great value in settling ques- 
 tions relating to valency and atomic weight, has been 
 obtained by the study of these compounds. 
 
 The following are the formulae of some of the most 
 interesting organo-metallic bodies. It will be noticed that 
 they are constituted in the same way as the chlorides and 
 oxides of the same hydrocarbon radicles, the special pe- 
 culiarities of the organo-metallic bodies being due to the 
 unoxidised condition of the metals they contain. 
 
 The vapour densities of all these compounds are normal, 
 so that each formula represents two volumes of vapour. 
 
 Ethyl chloride . . r\ r 11 
 or chlorine ethide . . ^ 2t5 
 
 Ethyl oxide . . . ft{C 2 H 5 
 or oxygen ethide . . (C 2 H 5 
 
Organo.-metallic Compounds. 
 
 Boron ethide . 
 
 C 2 H 5 
 
 333 
 
 Silicon ethide . 
 
 Trimethyl arsine 
 
 or arsenious trirnethide 
 
 Diarsemous tetramethide 
 (kakodyl) 
 
 Trimethyl stibine 
 
 or antimonious trimethide 
 
 Triethyl bismuth ine 
 or bismuthous triethide 
 
 Zinc ethide 
 Mercuric ethide 
 Stannous ethide 
 
 Stannic ethide 
 Plumbic ethide 
 
 ,,,(C 2 H a 
 
 B C 2 H 5 
 
 iv I f~* "H" 
 Cj ' ( -2 hL 5 
 
 Sl C 2 H 5 
 
 1 C* "W 
 V \^2 -^5 
 
 ,i/).CH a 
 
 As CH 3 
 )CH 3 
 
 /ACH 3 ) S 
 
 Sb 
 
 
 (As(CH 3 ) 2 
 
 CH 3 
 CH 3 
 CH, 
 
 Bi C,H fl 
 lc 2 H 
 
 : ; ;(C a H 5 
 
 Sn 
 
 SnJ C H 
 
 b C 2 H 5 
 
 (C 2 H 5 
 (C 2 H 5 
 
 iv p TJT 
 
 pK ) '-2 rl 5 
 
 Pb C.H, 
 
 1C 2 H 5 
 
334 Chemical Philosophy. 
 
 EXERCISES ON SECTION VI. 
 
 1. Write down the formulae of all the possible silver salts of the 
 following acids : 
 
 HC10 3 , HC10 4 , H,S0 4 , H 3 P0 4 , H,SiO 4 , H 2 SO 3 , H 8 PO 3 . 
 
 2. Write down the formulae for all the chlorides and amides theoreti- 
 cally derivable from the following acids : 
 
 UNO., H 2 SO 4 , H 4 P 2 O., HC 7 H 5 O, (benzoic acid, monobasic). 
 
 3. Write the constitutional formulae of Nordhausen sulphuric acid, 
 H 2 S 2 O 7 , of permanganic acid, H 2 Mn_O s , and of chromic acid, H 2 CrO 4 , 
 assuming sulphur, manganese, and chromium hexad. 
 
 4. Write down the formulas for the possible anhydrides derived from 
 orthosulphuric acid, S(OH) d , by successive stages of dehydration. 
 
 5. Give the formulae for the anhydrides corresponding with sulphuric 
 acid, H 2 SO 4 , phosphoric acid, H 3 PO 4 , pyrophosphoric acid, H 4 P_O 7 , 
 nitric acid, HNO 3 , nitrous acid, HNO 2 , hyponitrous acid, HNO, 
 theiosulphuric acid, H.^S.^Og, chloric acid, HC1O 3 , perchloric acid, 
 HC1O 4 , and acetic acid, HC 2 H 3 O 2 , pointing out those which are actually 
 known. 
 
 \,/~ 6. Give concise definitions of the terms acid, base, salt, with ex- 
 amples of their characteristic properties drawn from your own expe- 
 rience. 
 // 7. How would you explain the terms acid, and neutral salt ? Give 
 
 / examples. 
 
 8. -A number of salts having the general formula M'HSO 2 were 
 called by the discoverer hydrosulphites. What name should they bear 
 according to the rules laid down in Chap. XIII. ? Write the graphic 
 formulae for these bodies, and mention compounds which are similarly 
 constituted. 
 
 9. Complete the following equations, adding the name to each 
 formula : 
 
 HNO 3 + PbO = 
 HNO 8 + NH 3 = 
 H 2 SO 4 + KHO = 
 H.,S0 4 + PC1 5 = 
 (NH 4 ),S0 4 - 2H.O = 
 (NHJliSO, - ILO = 
 (NH 4 ) 2 CO, - H,6 = 
 (NH 4 )',CO 3 -2lip = 
 C 2 H.O"C1 + OH 2 = 
 Ca(HO) 2 -H,o" = 
 
Exercises on Section VL 335 
 
 CaCl 2 + Na,CO 8 = 
 CaCO 3 + 2NH 4 C1 = 
 NH, + C 2 H 5 I = 
 NH,C 2 H 5 + C,H,I = 
 NH(C,H 3 ) 2 + C..H.I => 
 N(C 2 H,) a + C,H,I = 
 N(C 2 H 5 ) 4 I + AgHO = 
 
 10. Give reasons in favour of each of these formulae for sal- 
 ammoniac : 
 
 NH 3 . II Cl 
 NH 4 C1 
 
 11. To which class of compounds would you refer the following 
 substances ? 
 
 Caustic potash, KIIO ; lime, CaO ; nitrous oxide, N.X) ; nitrogen 
 trioxide, N 2 O 3 ; ferric oxide, Fe.,O 3 ; chromium trioxide, CrO 3 ; wood 
 spirit, CH 3 .OH ; glycerine, C 3 H 5 (OH) 8 ; urea, CO(NH J ), ) . 
 
 12. Accepting the formula HC1 for hydrochloric acid, what facts 
 determine the choice of the formula H,C,O 4 for oxalic acid ? 
 
 13. How far do the following compounds agree with definitions of 
 the terms acid, base, salt ? Give sufficient reasons in each case. 
 
 Chlorides of hydrogen, potassium, aluminium, phosphorus, sulphur ; 
 oxides of hydrogen, potassium, carbon ; hydroxides of potassium, alu- 
 minium, arsenic, silicon. 
 
 14. Define the terms hydrate and hydroxide. Classify the following 
 compounds : 
 
 Cl,.ioH 2 O, KHO, H 3 A1O S , C,H 5 OH. 
 
 15. Write the constitutional formula of urea, and of the compound 
 with which it is isomeric. 
 
 1 6. How many litres of marsh-gas would be equal in weight to 25 
 litres of ethylene ? 
 
 17. Benzoate of silver has the formula C ; H 5 AgO 2 . Calculate its 
 percentage composition, also the weight of silver contained in "5736 
 gram of this salt. 
 
 18. From the formulae HBr, ILS, CC1,, NH 4 C1, H,SO 4> state the 
 valence of the symbols Br, S, C, N, and S, giving your reasons in 
 each case. 
 
 .19. Name and classify the following compounds : 
 CH 4 , CHC1 3 , C.,H 6 
 
 C 2 H 5 C1, C 2 H 5 
 
 CH 3 ) CH, 
 
336 Chemical Philosophy. 
 
 20. Assign a formula to a body containing 26-6, H2-2, and O;i -I 
 per cent. 
 
 21. How would you prove that the formula of marsh-gas is CH 4 ? 
 Also, how could you show that ethylene contains twice as much carbon ? 
 
 22. Give reasons for representing oxalic acid as C 2 H.,O 4 , not CHO 2 ; 
 acetic acid as C1I,O_,, not CH 2 O ; ethylene as C 2 H 4 , not CH 2 . 
 
 23. By what experiments could you prove that the molecule of ben- 
 zene is correctly represented by the formula C ti H 6 , not C 3 H 3 ? 
 
 24. A body having the composition C 5 H )2 O is an alcohol. Express 
 by equations the probable action upon it of acetic acid, oxalic acid, and 
 oxidising agents. 
 
 25. The compound C 4 H S O is either an aldehyd or a ketone : how 
 would you recognise its true character ? Write out the constitutional 
 formula of the aldehyd and ketone of this composition, and indicate 
 the constitution of any me tarn eric modification that may exist. 
 
 26. Dissect the crude formula C 7 H I4 (X, showing the nature of the 
 several metameric compounds which it may represent. 
 
 27. In respect of what properties may you regard alcohols as water 
 molecules in which hydrogen has been replaced by hydrocarbon radi- 
 cles ? Illustrate by reference to common alcohol, C-jH^OH. 
 
 28. The ethylic series of monohydric alcohols are represented by the 
 general formula 
 
 CaHja+jOH. 
 
 Write out the formula for the corresponding aldehyds, acids, and acetic 
 derivatives (acetates). 
 
 29. How many acids could be derived from the alcohol 
 
 CH 2 .OH, 
 
 and what would be their formulae ? 
 
 30. Carbolic acid is sometimes called phenylic alcohol. In what 
 respects do its alcoholic properties differ from those of the alcohols of 
 the ethylic series, C n H 2n +2 O ? 
 
 31. Complete these equations : 
 
 CH 3 .CO(OH) + PC1 5 - 
 CH 3 .OH + HC1 = 
 
 2 CH 3 .CO(OH) + Ag 2 O 
 
 CH,.CO(OCH 3 ) + NaHO 
 CH,.CO ' + H ' 
 
Exercises on Section VI. 337 
 
 32. In what respect does prussic acid resemble and how does it 
 differ from hydrochloric and hydrobromic acids ? 
 
 33. "3339 gram of a compound gave 7896 gram of CO., and -2924 
 gram H 2 O. The vapour density (air = i) was found to be 6 '59- 
 Calculate a rational formula. 
 
 34. '1483 gram of ethylated acetone gave '3799 gram of CO 2 and 
 1575 gram of water. Its vapour density (air = i) was 2-951. Calcu- 
 late the formula of the compound. 
 
 35. The analysis of barium diacetotartrate gave the following 
 results : 
 
 I. -2065 gram of substance gave -1297 gram of BaSO 4 . 
 n - ' r 377 gram of substance gave -1307 of CO 2 and -0300 of H 2 O. 
 
 Calculate the formula of the salt and of the corresponding acid. 
 
 36. The following results were obtained in the analysis of the copper 
 salt of ethyl-benzoic acid : '2218 gram gave '4827 gram of CO 2 , '1004 
 gram of water, and '0489 gram of CuO. Calculate the formula of the 
 acid. 
 
INDEX. 
 
 ABS CHE 
 
 A BSOLUTE tempe- 
 
 Chemical affinity, 5, 203 
 
 .f~X ratures, 51 
 
 combination, laws of, 
 
 Absorption spectra, 59 
 Acids, 296-328 
 
 79 
 compounds, characters 
 
 Adhesion and chemical 
 
 of, 92 
 
 affinity, 209 
 
 Chemistry, province of, 
 
 Affinity, 5, 203 
 
 5 
 
 Alcohols, 321 
 
 Chlorine, 220 
 
 Aldehyds, 325 
 Alkaline earths, 261 
 
 Chromium, 273 
 Classification of carbon 
 
 Alkalis, 259 
 
 compounds, 314 
 
 Allotropy, 199 
 Aluminium, 272-273 
 
 of compounds, 296 
 of elements, 217 
 
 Amides, 312 
 
 of reactions, 89 
 
 Amines, 308 
 
 Cobalt, 273 
 
 Ammonia, derivatives of, 
 
 Coefficient of absorption 
 
 307-33 
 
 of gases, 27 
 
 Ammonium theory, 307 
 
 Cohesion, 7 
 
 Analysis and synthesis, 5 
 
 Colloids, 30 
 
 Antimony, 250 
 
 Combustion, 186 
 
 Arsenic, 250 
 Arsines, 312 
 
 Compound ethers, 331 
 radicles, 148 
 
 Asymmetric carbon, 197 
 
 Compounds and mix- 
 
 Atomic theory, 85 
 - - weights, 109 
 
 tures, 92 
 Conditions of chemical 
 
 Atomicity, 140 
 Atoms, 4 
 
 change, 161 
 Constitution distin- 
 
 Avidity, 213 
 
 guished from composi- 
 
 BARIUM, 261 
 Bases, 296-303 
 Basic salts, 305 
 Basigenic elements, 218 
 Benzene, constitution of, 
 318 
 
 tion, 6 
 Constitutional formula;, 
 144, 197, 198 
 Copper, 273, 279 
 Critical point, 10, 38 
 Cryohydrates, 16 
 Crystallisation from solu- ' 
 
 Bismuth, 250 
 Boiling points, 34 
 
 tions, 25 
 Crystalloids, 30 
 
 Boron, 235 
 Boyle's law, 46 
 Bromine, 220 
 
 D ALTON'S atomic 
 theory, 83 
 Davy on flame, 64 
 
 
 Davy's experiments on 
 
 CADMIUM, 262 
 
 chlorine, 71 
 
 V_x Caesium, 259 
 
 Dialysis, 31 
 
 Calcium, 261 
 
 Diffusion of gases, 40 
 
 Carbon, 236 
 
 of liquids, 29 
 
 compounds, 314 
 Catalytic actions, 165 
 
 Dimorphism, 193 
 Dissociation, 168 
 
 Charles's law of expan- 
 
 Dumas on atomic weights, 
 
 sion of gases, 48 
 
 85 
 
 GAS 
 
 EBULLITION, 33 
 Efflorescence, 170 
 Electricity and chemical 
 
 action, 183 
 Electrochemical theories, 
 
 205 
 
 Electrolysis, 184, 205 
 Elements and compounds, 
 70 
 
 basigenic, 218 
 
 non-metallic, 217 
 
 oxygenic, 217 
 
 table of, 76 
 
 Energy and chemical 
 
 change, 6, 176 
 of isomerides, 202 
 Equations, 86 
 Ethereal salts, 331 
 Ethers, 325 
 Eutectic mixtures, 15 
 Exercises on Section I., 67 
 
 on Section II., 105 
 
 on Section III., 153 
 
 on Section IV., 215 
 
 on Section V., 293 
 
 on Section VI., 334 
 
 FLAME, luminosity 
 of 6 4 
 
 structure of, 65 
 
 temperature of, 188 
 Fluorine, 220-224 
 Formula explained, 78 
 Fractional distillation, 36 
 Frankland on flame, 65 
 Fraunhofer's lines, 60 
 Fusion, 12 
 
 r* ALLIUM, 272 
 VJ Gases, 7, 
 
 diffusion of, 40 
 
 expansion of, by heat, 
 
 -1 
 
 liquefaction of, 36 
 
 occlusion of, 44 
 
 passage of, through 
 metals and membranes, 
 44 
 
 pressure of, 46 
 
 solution of, 26 
 
340 
 
 Chemical Philosophy. 
 
 GAY 
 
 MET 
 
 ZIR 
 
 Gay-Lussac's law of ex- 
 
 Metalloids, 244 
 
 ' Selenion, 227 
 
 pansion, 48 
 
 Metals, 256 
 
 Silicon, 236 
 
 of volumes, 82 
 
 Metamerism, 195 
 
 Silver, 265 
 
 Germanium, 248 
 
 Metric weights and mea- 
 
 Sodium, 259 
 
 Gold, 281 
 
 sures, 2 
 
 Solids and fluids, 6 
 
 Graham's law of diffu- 
 
 Meyer's curve of atomic 
 
 Solution of gases, 26 
 
 sion, 41 
 
 weights and volumes. 
 
 of solids, 1 6 
 
 Graphic formulae, 144 
 
 289 
 
 theories of, 24 
 
 
 Mixtures and compounds, 
 
 Solutions,supersaturated, 
 
 HALOGENS 220 
 
 96 
 
 26 
 
 Heats of combus- 
 
 Molecular compounds, 
 
 Solvents, influence of, 162 
 
 tion, 189 
 
 211, 227 
 
 Spectra, 53 
 
 of formation of chloi 
 
 theory, 8 
 
 absorption, 59 
 
 rides, &c., 180 
 
 weights and formula;, 
 
 emission, 53 
 
 of neutralisation of 
 
 122 
 
 invisible, 63 
 
 acids, 178 
 
 Molecules, 3 
 
 Stibines, 312 
 
 Homologous series, 35 
 
 of elements, 123 
 
 Strontium, 261 
 
 Hydrocarbons, 315 
 
 Molybdenum, 255 
 
 Substitution compounds, 
 
 Hydrogen, 245 
 
 
 133 
 
 TNDIUM, 272 
 
 NAMES of com- 
 pounds, 101 
 
 Sulphur, 227-234 
 
 JL Iodine, 220-222 
 
 of elements, 100 
 
 TANTALUM, 254 
 
 Iridium, 282 
 
 Nascent state, 126 
 
 Tellurium, 247 
 
 Iron, 273 
 
 Newlands on atomic 
 
 Thallium, 268 
 
 Isomerism, 192 
 
 weights, 288 
 
 Thermal effects of ther- 
 
 Isomorphism, 116 
 
 Nickel, 273 
 
 mical action, 176 
 
 
 Niobium, 254 
 
 Thermolysis, 169 
 
 T^ETONES, 325 v 
 
 Nitrogen, 239 
 
 Tin, 248 
 
 1\. 
 
 Nomenclature, 99 
 
 Titanium, 248 
 
 
 Non-metals, 219 
 
 Tungsten, 254 
 
 LAW of Avogadro, 51 
 of Boyle, 46 
 
 OCCLUSION of 
 
 Types, 136 
 
 of Charles, 48 
 of chemical combina- 
 
 gases, 44 
 Organo-metallic bodies, 
 
 T TRANIUM, 255 
 
 tion, 79 
 
 33 2 
 
 
 of diffusion Q/ gases, 41 
 ofDulongand f etit,ii4 
 of even numbers, 146 
 
 Osmium, 282 
 Oxygen, 227 
 Ozone, 231 
 
 VALENCY, 139 
 of carbon and 
 silicon, 238 
 
 of Gay-Lussac (law of 
 volumes), 83 
 of isomorphism, 116 
 of volumes (Gay-Lus- 
 sac), 83 
 Lead, 268 
 Liquefaction of gases, 38 
 Liquid diffusion, 29 
 Liquids, 7 
 Lithium, 259 
 Luminosity of flame, 64 
 
 PALLADIUM, 282 
 Periodic law, 284 
 Phlogiston, 187 
 Phosphines, 312 
 Phosphorus, 239-241 
 Physical isomerism, 192 
 Platinum, 282 
 Polymerism, 194 
 Potassium, 259 
 Prout's hypothesis, 284 
 
 of common salt radicles, 
 table, 152 
 of oxygen and sulphur, 
 
 of principal elements, 
 143 
 of the halogens, 225 
 of the .iron group, 277 
 of tin and its allies, 247 
 Vanadium, 250 
 Vapour densities, abnor- 
 
 MAGNESIUM, 262 
 Manganese, 273 
 Mass, action of, 163 
 Matter, constitution of, 3 
 
 RATIONAL formulae, 
 137 
 Reactions classified, 89 * 
 Rhodium, 282 
 Rubidium, 259 
 
 mal, 171 
 density, 123 
 
 WATER of crystalli- 
 sation, 25 
 
 Melting points of mix- 
 tures, 14 
 
 Ruthenium, 282 
 
 Weights and measures, 
 table of, 2 
 
 MendelejefFs classifica- 
 tion of elements, 291 
 Mercury, 265 
 
 SALTS, 296-305 
 Saturating power of 
 acids and bases, 131 
 
 st^tfk 
 
 ZINC, 262 
 Zirconium, 248 
 
 ^0^ 
 
 Spottiswoodt & Co. Prini 
 
THIS BOOK IS DUE ON THE LAST DATE 
 STAMPED BELOW 
 
 AN INITIAL FINE OF 25 CENTS 
 
 WILL BE ASSESSED FOR FAILURE TO RETURN 
 THIS BOOK ON THE DATE DUE. THE PENALTY 
 WILL INCREASE TO SO CENTS ON THE FOURTH 
 DAY AND TO $1.OO ON THE SEVENTH DAY 
 OVERDUE. 
 
 APH.l 1S42 
 
 1 9 
 
 3,9