MEMCAL 
 
 COLLEGE OF 
 
 v\ 
 
W ; T. WENZELl 
 Francisco, OL 
 
 INTEODUCTION 
 
 TO 
 
 MODEEN CHEMISTRY 
 
 EXPERIMENTAL AND THEORETIC. 
 
 EMBODYING 
 
 TWELVE LECTUEES DELIVERED IN THE ROYAL COLLEGE OF 
 CHEMISTRY, LONDON, 
 
 BY 
 
 A. W. JEOFMANN, LL.D., F.R.S., V.P.C.S. 
 
 OR OP CHEMISTRY IN THE ROYAL SCHOOL OF MINES; ASSAYER TO THE ROYAL MINT 
 ER IN CHEMISTRY TO THE DEPARTMENT OF SCIENCE AND ART. 
 
 rrJa CoISego of Pharmaf 
 
 LONDON: 
 WALTON AND MABEELY, 
 
 aOWER STREET, AND IVY LANE, PATERNOSTER ROW. 
 
 1806, 
 
LONDON: PRINTED EY WILLIAM CLOWES AND SONS, STAMFORD STREET 
 AND CHARING CROSS. 
 
ffVf 
 
 TO 
 
 SIR JAMES CLARK, BART., M.D., F.R.S. 
 
 PHYSICIAN TO HER MAJESTY THE QUEEN. 
 
 MY DEAR SIR JAMES, 
 
 Two motives, either of which would be adequate, inspire 
 me with the wish to dedicate this work to you. 
 
 In the first place, the lectures it embodies were delivered 
 in an Institution of which, under the auspices of the illustrious 
 Prince Consort, you were one of the principal founders, and 
 which owed its passage through the perils of infancy mainly to 
 your unremitting and strenuous support. 
 
 In the second place, appearing, as this book does, at a 
 moment when duty calls me away to labour in another sphere, it 
 affords a particularly fitting occasion for the acknowledgment of 
 my deep debt of gratitude to one whose sympathy has so kindly 
 and constantly sustained me in endeavouring to promote, in the 
 country of my adoption, the great cause of chemical education. 
 
 Not for any intrinsic merit of its own, but as one effort more 
 in that noble cause, you will accept, I am persuaded, my dear 
 Sir James, the dedication of this little work in token of my 
 sincere regard and unalterable friendship. 
 
 A. W. HOFMANN. 
 
 London, March 25, 1865. 
 
 427 
 
>> ^ofPharmac 
 
 PEEFACE. 
 
 THE following pages contain, in their latest form, with some 
 additional developments, the introductory portion of the chemi- 
 cal course which the Author has annually delivered, during 
 the last fifteen years, in the theatre of the Eoyal College of 
 Chemistry. 
 
 The juncture of chemical history at which this little book is 
 published, and the peculiar educational necessities which that 
 juncture implies, are reflected as well in the substance as in the 
 method of the work ; the purport of which will, therefore, be 
 best made known by a brief reference to the present posture of 
 chemical affairs. 
 
 No chemist will need to be reminded that, during the last 
 quarter of a century, the science of chemistry has undergone a 
 profound transformation ; attended, during its accomplishment, 
 by struggles so convulsive, as to represent what, in political 
 parlance, would be appropriately termed a Revolution. 
 
 Amidst continual accessions of fact, so rapid, so voluminous, 
 and so heterogeneous, as almost to exceed the grasp of any single 
 mind, chemical science has been in travail, so to speak, with 
 new laws and principles of co-ordination, engendered, perhaps, 
 partly by the sheer force of their own deeply-felt necessity, but 
 partly also, and mainly due, to the powerful initiative impulsion 
 of a few philosophical master-minds. 
 
 Based on the concurrent examination of the volumetric and 
 
VI PREFACE. 
 
 ponderal combining-ratios of certain typical elements, and on 
 the recognition, in their standard combinations, of a few well- 
 marked structural types, these principles have introduced into 
 the domain of chemistry the pregnant idea of Classification 
 the conception of a series of natural Groups, resembling the 
 genera of the biological sciences, and culminating in the esta- 
 blishment of an orderly System, where before there had seemed 
 to be but a chaos of disconnected facts. 
 
 Under the influence of these and certain other cognate ideas, 
 new views have arisen as to the constitution and chemical pro- 
 perties of matter ; a reformed chemical notation has thence of 
 necessity ensued ; and structural relations, previously unsus- 
 pected, have disclosed identity of parentage in compounds till 
 then deemed utterly diverse. 
 
 It appears to be wisely ordered, in scientific as in social 
 affairs, that the innovating spirit which belongs to Youth has 
 its check and counterpoise in the conservative tendency essen- 
 tially characteristic of Age ; so that, in the sharp collision of 
 these rival forces, new principles, in any kind, find a sort of 
 fiery ordeal interposed between their first enunciation and final 
 acceptance ; doubtless the appointed test of their soundness and 
 vitality. 
 
 Hence the domain of chemical philosophy has, for many years 
 past, rather resembled a tumultuous battle-plain, than a field 
 bestowed by nature for peaceful cultivation by mankind. The 
 new ideas, springing up of necessity one by one, and not always 
 free, at their first conception, from errors and inconsistencies, 
 have been resisted, by the champions of the old chemical dogmas, 
 as a gratuitous revolt against established authority. Controversy 
 has naturally stimulated research, which, in its turn, has produced 
 rapid modifications of theory ; so that the aspect of chemistry 
 has been in a state of incessant change. It is, indeed, only 
 within the last few years that the new doctrines have acquired a 
 
PREFACE. Vll 
 
 logical consistency, and a consequent ascendency throughout 
 Europe, auguring at length, for our long-agitated science, a 
 period of comparative calm. 
 
 The Author's chemical lot, both as a student and as a teacher, 
 has been cast amidst the storms of this controversial period ; in 
 which he has felt it his duty to take part on the side of inno- 
 vation. 
 
 During the lengthened period which he has passed in tuition at 
 the College of Chemistry, the task imposed on him has been, to 
 hold the balance fairly between the old views and the new ; be- 
 tween authorized conceptions, evidently on the wane, and novel 
 generalizations still awaiting final proof ; so that his teachings 
 may have sometimes almost seemed to resemble those dissolving 
 scenes, which, at a certain moment, present two landscapes, one 
 in the act of melting away, while the other is unfolding itself to 
 view. 
 
 It will be readily understood that a printed record of such 
 discourses, though frequently and earnestly solicited by pupils, 
 could scarcely have possessed any permanent value, had it been 
 produced amidst the doubts, and reflected the half-truths, of a 
 period so eminently transitional. It will also be felt that its 
 publication has become more opportune, now that a consistent 
 body of doctrine, novel, yet based on irrefragable fact, can be 
 put forth as undoubtedly permanent. 
 
 It will be apparent from the above remarks that this work is 
 of an essentially general and introductory character, designed to 
 elucidate the leading principles of chemistry, and by no means 
 presented as an encyclopaedic compendium of its facts. So far, 
 indeed, from seeking to multiply details, it has been the author's 
 chief care to avoid them, and to enter upon descriptions of 
 phenomena only in subordination to his main design. This 
 will, perhaps, be most readily gathered from the brief resume 
 of these lectures, with which the last of the series concludes. 
 
Vlll PEEFACB. 
 
 Moreover, as the Author's aim is essentially educational, so 
 he has made it his care to avoid a too systematic treatment of 
 his subjects ; preferring the experimental, illustrative method, 
 so peculiarly adapted for the lecture-room. Thus, at the very 
 outset, he plunges in medias res ; starting abruptly with an ex- 
 periment ; and leaving the principles therein involved to unfold 
 themselves naturally from the explanation of the phenomena 
 displayed. Throughout the course he has adhered to this 
 method ; proceeding constantly from the concrete to the general ; 
 and extracting from a limited range of facts the largest amount 
 of theoretic and general information which they can readily be 
 made to yield. 
 
 In the fulfilment of this programme he has had of necessity to 
 break with the classical traditions of chemical teaching. The 
 elements, for example, are here studied in a new order, not 
 gratuitously adopted, but determined by the Author's view of 
 their fitness to lead up to the knowledge of general laws, in just 
 and logical succession ; and so also, while endeavouring to illus- 
 trate, incidentally, the leading topics of experimental chemistry, 
 he has been mainly guided, in his selection of experiments, by 
 their fitness for the elucidation of theoretical views. 
 
 It is rather in accordance with these exigences of his plan, 
 than with reference to the relative importance of the subjects 
 treated, that space has been meted out to these ; some topics, in 
 themselves of great moment, being dismissed with but a passing 
 notice ; while others, intrinsically less interesting, are elaborately 
 discussed, on account of their bearing on questions of principle. 
 
 The execution of such a project as this is greatly facilitated 
 by the Lecturer's happy prerogatives. The mere limits of time 
 to which he is bound, preclude, in any case, his attempting the 
 exhaustive treatment of his themes. At the lecture-table he is 
 only expected to display a few salient facts, in a striking and 
 attractive form, and to deduce therefrom a few guiding princi- 
 
PREFACE. IX 
 
 pies, so as to assist his auditors in acquiring for themselves the 
 details of the science. Any attempt on the lecturer's part to 
 make his brief discourses encyclopedic must, of necessity, fail ; 
 nay, lectures are probably by so much the better fitted for their 
 purpose, by how much they are freer from unnecessary detail, 
 and more thoroughly emancipated from the trammels of sys- 
 tematic routine. 
 
 The Author touches on these points in order that more 
 may not be expected from this book than its very design 
 permits it to supply. Its text, indeed, as originally written, 
 was an almost verbatim reproduction of his actual language 
 at the lecture-table ; the experiments described are those which 
 were really shown before the class ; and the woodcuts faith- 
 fully represent the apparatus employed in their performance.* 
 
 Into the original framework thus provided, there has been 
 introduced, in preparing these discourses for the press, such ad- 
 ditional matter as appeared desirable for the more complete 
 elucidation of the great laws set forth. 
 
 In the performance of this task, which came upon him amidst 
 the almost overwhelming pressure involved in the simultaneous 
 completion of old engagements and preparation for new ones, 
 the Author gladly availed himself of the kindly proffered col- 
 laboration of his esteemed friend, Mr. F. O. Ward ; whose well- 
 known powers of lucid composition, and habits of philosophical 
 thought, will be traced in every chapter of this work. Attracted 
 to the new chemical doctrines by their own intrinsic truth and 
 beauty, Mr. F. 0. Ward has willingly devoted himself, for 
 months past, to the task of assisting in their exposition ; and in 
 
 * The Author cannot refrain from acknowledging the great pains be- 
 stowed by the artist, Mr. Julius Jury, in delineating these illustrations ; and 
 by the engraver, Mr. Rexworthy De Wilde, in executing them. His thanks 
 are also due to Messrs. McLeod and Oilman, for their care and skill in 
 making the photographs from which many of the drawings were copied. 
 
X PREFACE. 
 
 the course of these labours, as was to be expected, he has 
 originated many valuable conceptions for their clearer elucida- 
 tion and development. One, indeed, of his friend's indications 
 the Author feels bound to mention here, as constituting a dis- 
 tinct and valuable contribution to the new chemical edifice. 
 He alludes to the Quantivalential Equilibrium of the Nitroxy- 
 gen series, as demonstrated by Mr. Ward, and displayed in the 
 striking symmetrical diagram introduced by him at p. 183. 
 
 In conclusion, the Author desires to express his heartfelt 
 thanks to Mr. F. 0. Ward for his invaluable collaboration. 
 
 Royal College of Chemistry, London. 
 March 24, 1865. 
 
CONTENTS. 
 
 LECTURE I. 
 
 PAGE 
 
 Water its decomposition by the alkali-metals the resulting gas, 
 hydrogen its principle characters its volume- weight other sources 
 of hydrogen muriatic acid and ammonia dissolved in water 
 disengaged as gases desiccation of the gases their distinctive 
 properties development of hydrogen from muriatic acid and am- 
 monia by the alkali-metals processes and apparatus further sources 
 of hydrogen decomposition of muriatic acid, water, and ammonia, 
 and disengagement of hydrogen therefrom by electric agency . . 1 
 
 LECTURE II. 
 
 Electrical investigation of muriatic acid, water, and ammonia action 
 of the electric current on muriatic acid development therefrom of 
 mixed hydrogen and chlorine separation of chlorine from the 
 mixture distinctive properties of chlorine reproduction of muriatic 
 acid, by synthesis, from hydrogen and chlorine whence its modern 
 name hydrochloric acid action of the electric current on water 
 development therefrom of mixed hydrogen and oxygen separation 
 thereof distinctive properties of oxygen liberation of oxygen from 
 water by chlorine synthetical reproduction of water from hydrogen 
 and oxygen action of the electric current upon ammonia evolution 
 therefrom of mixed hydrogen and nitrogen separation thereof 
 distinctive properties of nitrogen its liberation from ammonia by 
 means of chlorine synthesis of ammonia from its elements not yet 
 achieved proofs of the composition of ammonia matter, simple and 
 compound table of elements 15 
 
XU CONTENTS. 
 
 LECTURE HI. 
 
 PAGE 
 
 Compound bodies their volumetric constitution and condensation 
 ratios as exemplified in hydrochloric acid in water in ammonia 
 illustrative experiments mechanical mixture and chemical combi- 
 nation their distinguishing characteristics experimental illustra- 
 tions thereof mixtures and combinations of the elements of hydro- 
 chloric acid of water constancy of chemical composition marked 
 changes of property attending chemical combination . . . 39 
 
 LECTURE IV. 
 
 Chemical symbols their nature and value diagrammatic symbols 
 initialed figured symbolic equations constructed therewith infor- 
 mation thereby conveyed formulae thence derived summary of 
 information condensed in chemical formulae hydrochloric acid, 
 water, and ammonia, considered as types of chemical combination 
 hydrobromic and hydriodic acids their construction upon the type 
 of hydrochloric acid ponderal analysis and volume-weights of these 
 compounds volume-weights of bromine and iodine gases sulphur- 
 etted and selenetted hydrogen their construction upon the type of 
 water ponderal analysis and volume- weights of these compounds 
 volume-weights of sulphur and selenium gases 67 
 
 LECTURE V. 
 
 Chemical symbols (continued) phosphoretted and arsenetted hydrogen 
 their construction upon the type of ammonia ponderal analysis and 
 volume-weights of these compounds exceptional volume-weights of 
 phosphorus and arsenic gases combining weights of phosphorus 
 and arsenic general remarks on chemical symbolization chemical 
 formulse as instruments of classification representation of chemical 
 processes in equations translation of formulse into concrete weights 
 and volumes ponderal analysis of sodic and potassic chlorides, 
 oxides, and nitrides determination of the combining weights of 
 sodium and potassium 80 
 
 LECTURE VI. 
 
 Fourth term in the series of typical hydrogen compounds marsh-gas, 
 or light carbonetted hydrogen reasons for its separate consideration 
 its occurrence in marshes in coal-mines in coal-gas its pre- 
 parationits distinctive characters its qualitative analysis separa- 
 tion therefrom of carbon by chlorine decomposition of marsh-gas 
 by heat its quantitative analysis its synthesis not yet accomplished 
 combining weight of its constituent, carbon its analogues sili- 
 cetted hydrogen probability of the construction thereof upon the 
 type of marsh-gas 98 
 
CONTENTS. Xlll 
 
 LECTUKE VII. 
 
 PAGE 
 
 Deportment of nitrogen towards oxygen nitric acid hydrated anhy- 
 drous its composition its decomposition by heat by metals by 
 tin, yielding hyponitric acid by silver, yielding nitrous acid by 
 copper, yielding nitric oxide by zinc, yielding nitrous oxide 
 characters of these products how shown to be chemical compounds, 
 not mechanical mixtures expansion of the idea of chemical combi- 
 nation combination of two elements in several proportions law of 
 multiple proportions volume and condensation ratios in chemical 
 compounds ordinary exceptional Ill 
 
 LECTUKE VIII. 
 
 Transition from abstract to concrete formulae choice of a system of 
 weights and measures to supply units for the expression of concrete 
 values obstacles thereto, as also to the diifusion of science gene- 
 rally, by the want of a universally-accepted system of weights and 
 measures French metrical system reasons for adopting it its 
 general characters principles of its nomenclature comparison with 
 English measures the hydrogen litre-weight, or crith the volume- 
 weights of elements or compounds = the absolute weights of 1 litre 
 at C C. and O m< 76 pressure, expressed in criths ..... 121 
 
 LECTUKE IX. 
 
 Philosophical conceptions of chemical phenomena hypotheses 
 theories matter its nature and essence its conditions, solid, fluid, 
 gaseous its activities, molar and molecular properties of molecules 
 their mutual cohesion and repulsion nature and properties of gases 
 their elasticity their latent heat nature of heat molecular 
 dynaspheres observed influence of temperature and pressure on 
 gases identical comportment of all gases under like variations 
 thereof composite structure of molecules conception of atoms 
 how arrived at final term of the known threefold divisibility of 
 matter its infinite divisibility, why not to be affirmed elementary 
 molecules diatomic molecules tetratomic molecules their sym- 
 bolization atomic and molecular forms of notation molecular sym- 
 bolization exemplified in the formulae of the nitroxygen series 
 comparative advantages of the atomic and molecular form of 
 notation 137 
 
 LECTURE X. 
 
 Molecular and atomic constitution of the typical compounds curious 
 relations, ponderal, numerical, and potential, of the typical elementary 
 
XIV CONTENTS. 
 
 PACK 
 
 atoms two sorts of chemical value or power, molecule-forming and 
 atom-fixingunitary standard of atom-fixing power major and minor 
 equivalent weights coefficients of atom-fixing power, or quantiva- 
 lence comparative quantivalence of the typical elements and their 
 congeners germ of a natural system of chemical classification 
 volume-condensing power of atoms, how far proportionate to their 
 quantivalence alternative standards of quantivalence chemical 
 value in exchange exemplification thereof in the syntheses of hydro- 
 chloric acid and water, and in the hypothetical syntheses of ammonia 
 and marsh-gas also in their decomposition by chlorine in the con- 
 trasted action of chlorine and oxygen on hydriodic acid and in the 
 comparative structure of ammonia and nitrous acid quantivalential 
 equilibrium of the nitroxygen series distinction between numerical 
 and potential quantivalence also between quantivalence and chemism 
 transitional state of the question tabular summary . . .163 
 
 LECTUEE XI. 
 
 Principles established for 13 elements, how far applicable to the re- 
 maining 48 method of determining the atom-weights of elements, 
 neither volatile themselves, nor capable of forming volatile com- 
 pounds with hydrogen value of volatile chlorine-compounds for this 
 purpose also of volatile oxygen and nitrogen compounds when such 
 exist illustrations of the principle supplied by such as are volatile of 
 the chlorine, oxygen, and nitrogen compounds of hydrogen, sodium, and 
 potassium atom- weights and quantivalence of sodium and potassium 
 as determined by the balance prospect of volumetric corroboration 
 of these results further illustrations mercury bismuth tin their 
 vapour-densities ponderal composition of their respective chlorides 
 deductions therefrom of their respective atom-weights and coeffi- 
 cients of quantivalence universal applicability of this method its 
 liability to error in one respect -further data requisite to obviate this 
 such data how to be obtained curious exceptional relation of the 
 atomic and molecular weights of mercury as also of cadmium 
 contrast presented by these metals in this respect to phosphorus and 
 arsenic contrasted symbolization of these bodies monatomic, dia- 
 tomic, tetratomic, and polyatomic molecules corresponding anoma- 
 lies in the condensing powers of the respective atoms of these bodies 
 physical aids to the determination of atom-weights specific-heat 
 method isomorphic method cases in which they are of special 
 value table of the elements, with their atomic and molecular symbols, 
 weights, volumes (so far as known) and coefficients of quantivalence 
 one exception to the universality of the symbols appeal, on this 
 subject, to scientific France 187 
 
CONTENTS. XV 
 
 LECTURE XII. 
 
 PAGE 
 
 Compounds of a higher order, ternary, quaternary, &c. laws of propor- 
 tionality and quantivalential relations exemplified in the generation 
 of such compounds their frequently high vapour-densities their 
 tendency, in many cases, to dissociation examples of ternary com- 
 pounds hydrochlorate of ammonia its production by the union of 
 the molecules of two gaseous binary compounds its neutral, salt-like 
 characters dissociation of its vapour ternary compounds produced 
 during the progressive dehydrogenation of water and ammonia by 
 the alkali-metals unisodic water, or hydrosodic oxide unisodic and 
 bisodic ammonias, or hydro-unisodic and hydro-bisodic nitrides 
 progressive expulsion and replacement of hydrogen-atoms by sodium- 
 atoms during the formation of these compounds analogous substitu- 
 tion-compounds in the marsh-gas group general conception of sub- 
 stitution-compounds their production further illustrated in the 
 progressive transformation of water, ammonia, and marsh-gas, by 
 inception of chlorine and expulsion of hydrogen retention of the 
 structural type of the parent compound by its substitutional derivatives 
 conversion of binary into ternary compounds by atomic inception, 
 unattended by substitutional displacement exemplification of this 
 mode of their genesis in the hydrochloroxygen series analogous 
 ternary compounds of the hydrochloric group of the water group 
 hydrosulphoxygen series of the ammonia group hydrophosphoxy- 
 gen series of the marsh-gas group methylic alcohol its transi- 
 tional importance close of the lecture summary recapitulation and 
 conclusion of the course . 209 
 
INTRODUCTION TO MODEM CHEMISTRY, 
 
 EXPEEIMENTAL AND THEOEETIC. 
 
 LECTUKE I. 
 
 Water its decomposition by the alkali-metals the resulting gas, hydrogen 
 its principal characters its volume- weight other sources of hydrogen 
 muriatic acid and ammonia dissolved in water disengaged as gases 
 desiccation of the gases their distinctive properties development of 
 hydrogen from muriatic acid and ammonia by the alkali-metalsprocesses 
 and apparatus further sources of hydrogen decomposition of muriatic 
 acid, water, and ammonia, and disengagement of hydrogen therefrom by 
 electric agency. 
 
 IT is a well-known fact that Water can be brought into contact 
 with many of the metals, without undergoing any sensible 
 change. Gold and silver do not produce the slightest effect 
 upon it ; and even copper, iron, zinc, and tin, can be immersed 
 for a considerable time in water at the ordinary temperature of 
 the atmosphere, without causing any change in the condition of 
 the fluid. This does not, however, hold true of all metals. 
 
 Many metals produce a very decided action on water. 
 Amongst these we may, for 
 our present purpose, cite two in 
 particular, viz., Potassium and 
 Sodium, the one prepared from 
 wood ashes, the other from com- 
 mon salt, by processes to be ex- 
 plained hereafter. A ball of 
 potassium, thrown upon water 
 (Fig. 1), ignites, and glides with ^ 
 
 a hissing noise along the surface ; emitting a bright violet light 
 
WATER ITS DECOMPOSITION 
 
 evolving white fumes, and presently disappearing with a slight 
 report. A piece of sodium acts in a similar manner, though 
 the phenomenon is less striking. The floating ball does not 
 ignite unless its motion be arrested ; as, for instance, by spread- 
 ing a sheet of blotting-paper over the water, when the metal 
 burns with a bright yellow flame. In both cases the water 
 acquires a peculiar taste, called alkaline, and the faculty of modi- 
 fying vegetal colours not affected by pure water. A strip of 
 yellow turmeric paper, if dipped into water that has been exposed 
 to the action of potassium or sodium, turns brown, whilst red 
 litmus paper is changed to blue. 
 
 What becomes of the potassium and sodium which seem thus 
 to be consumed by the mere touch of water, and which in reality 
 cease to exist as metals ? 
 
 What, again, is the nature of that concomitant change in the 
 water itself, to which it owes its acquisition of a new and pecu- 
 liar flavour, and the power of transforming vegetal colours on 
 which it previously exerted no influence ? 
 
 These questions, and the class of modifications in the proper- 
 ties of matter to which they refer, belong to the science called 
 Chemistry, a term of obscure origin, traced by some to Xq/u'a, 
 FIG 2 one of the ancient names of 
 
 Egypt, where the study of these 
 mysterious transformations is 
 supposed to have taken its rise. 
 For the investigation of this 
 class of phenomena, which is 
 the purpose of these Lectures, 
 the striking facts we have just 
 witnessed, and the questions 
 which they suggest, furnish 
 an appropriate starting point ; 
 and I will therefore ask you 
 to accompany me in subjecting 
 them to a close and rigorous scrutiny. 
 
 For this purpose, let a cylindrical glass vessel, closed at 
 
BY THE ALKALI METALS. 3 
 
 one end, be filled with water, and its open mouth covered with 
 a flat glass plate; let it then be plunged niouth downwards 
 (Fig. 2) into a basin of water, and fixed in such manner that 
 its orifice may be beneath the surface of the water, without 
 touching the bottom of the vessel. The water will, of course, 
 be sustained by atmospheric pressure within the cylinder, so as 
 completely to fill it. Now let a ball of sodium be thrown 
 upon the water (potassium may be employed, but is less ap- 
 propriate on account of the greater violence of its action), and 
 conveyed, by means of a sort of spoon with a wire-gauze bowl, 
 beneath the mouth of the inverted cylinder (Fig. 3). Colourless 
 
 FIG. 3. 
 
 gas bubbles immediately appear, and rise in the cylinder, dis- 
 placing the water. By repeating this operation three or four 
 times the cylinder may be completely filled with gas. This done, 
 let the glass plate be replaced under the orifice (Fig. 4), and the 
 cylinder be raised out of the water and turned mouth upwards. 
 
 The gas which has been thus separated from the water, is 
 called Hydrogen. It is colourless, transparent, and has neither 
 taste nor smell; in these respects resembling common atmo- 
 spheric air, from which, however, it differs in many particulars. 
 
 B 2 
 
4 
 
 HYDROGEN ITS CHARACTERS, 
 
 Thus, for example, if brought into contact with a lighted candle 
 (Fig. 5), hydrogen inflames and bums with a pale lambent 
 flame, quietly descending into the vessel. 
 
 FIG. 4. 
 
 To secure this result, however, the glass plate must not be re- 
 moved from the mouth of the cylinder till this has been brought 
 
 FIG. 5. 
 
ITS VOLUME-WEIGHT ITS SOURCES. 
 
 close to the burning candle ; for if the cylinder be left open only 
 a few seconds, every trace of the inflammable gas will escape, 
 and only common air will remain in its place. If, on the other 
 hand, the glass plate be removed from the cylinder charged with 
 hydrogen, without turning the mouth upwards, the result will 
 be different. In this position twenty minutes or more may 
 elapse before the inflammable gas has dispersed ; a fact easily 
 ascertainable by applying the simple test of the lighted candle. 
 Instead of suffering the hydrogen thus to escape into the air, 
 it may be collected in a cylinder held mouth downward over 
 the ascending stream. In this way a jar full of hydrogen may, 
 so to speak, be poured or decanted upward, into an inverted 
 cylinder (Fig. 6). We infer from the facility with which the gas 
 escapes when an upward vent is afforded for its egress, that a 
 given bulk of hydrogen is lighter than an equal bulk of common 
 air. Accurate experiments have proved that the weights of equal 
 volumes of hydrogen and of common air are in the proportion of 
 1 to 14'4 ; in other words, that 
 the air is 14'4, or nearly 14^- 
 times heavier than hydrogen. 
 
 The weighing and measuring 
 of aeriform bodies involves cer- 
 tain precautions which we shall 
 have to consider more fully 
 hereafter ; for our present pur- 
 pose we need only remember that the bulk of an aeriform body 
 depends essentially on temperature and pressure, and that, 
 therefore, equality of these conditions is indispensable to any 
 just comparison of different gases in respect of their relative 
 volume-weight, or, as it is commonly termed, their specific 
 gravity. 
 
 Of all gases hitherto discovered, hydrogen is the lightest ; it 
 has, therefore, been deemed appropriate to take the weight of a 
 given bulk of hydrogen as the unit, in terms of which the weight 
 of an equal bulk of any other gas may be expressed. When, 
 therefore, we say that the specific gravity of the air is 144, we 
 
 FIG. 6. 
 
Q MUEIATIC ACID GAS J 
 
 mean that a given bulk of it, at given temperature and pressure, 
 weighs 14'4 times as much as an equal bulk of hydrogen under 
 like physical conditions. Hence the densities of gases are 
 necessarily and obviously proportionate to their specific gravi" 
 ties ; so that calling the density of hydrogen 1, the density of 
 atmospheric air, like its specific gravity, is 14-4. 
 
 Water, however, is not the only substance from which 
 hydrogen can be eliminated by means of potassium or sodium. 
 Two fluids, Muriatic acid and Ammonia, both of which have 
 been known for centuries past, and are at present extensively 
 employed in the arts, yield hydrogen when subjected to the action 
 of the metals above named. 
 
 Pure Muriatic acid is a gas, like common air, and like hydrogen. 
 The liquid muriatic acid of commerce is but an aqueous solution 
 of this gas, and gives it forth when heated. This operation is 
 conveniently performed in a glass flask, the mouth of which is 
 closed with a doubly perforated cork ; one of the perforations 
 carries a funnel-tube, the lower end of which dips into the 
 liquor, while its upper end widens out to a cuplike form, con- 
 venient for charging the vessel. Through the other perforation is 
 fitted a tube bent at right angles (elbow-tube), to give vent to the 
 gas evolved. 
 
 The gas issuing from this tube is, however, charged with 
 aqueous vapour, from which it must be freed by contact with a 
 desiccating agent; that is, with a substance more greedy of 
 moisture than itself. Such a substance is that known in 
 commerce as oil of vitriol, termed by chemists sulphuric acid. 
 This acid may be conveniently employed for the purpose of 
 desiccating gases. For this purpose a glass bottle, closed 
 by a doubly perforated cork, is provided. One of the perfora- 
 tions carries an elbow-tube, reaching to the bottom of the bottle ; 
 the other perforation is fitted with a deli very- tube. This bottle 
 being filled with pumice-stone saturated with sulphuric acid, and 
 its elbow-tube being attached by a piece of caoutchouc tubing, 
 called a connector, to the elbow-tube of the flask, the gas 
 issuing therefrom is conveyed to the bottom of the acid-soaked 
 
ITS CHARACTERS ITS DECOMPOSITION. 
 
 pumice-column, through the interstices of which it passes, im- 
 pinging as it rises against widely-spread surfaces of acid, which 
 greedily absorb its moisture. It thus reaches the upper part 
 of the bottle quite dry, and in that condition escapes from the 
 delivery-tube . 
 
 FIG. 7. 
 
 Muriatic acid vapours, thus treated, lose every particle of 
 water, and issue forth a colourless, transparent, and perfectly 
 pure gas. This may be collected in a cylindrical vessel filled 
 with mercury, inverted over a mercurial trough (Fig. 7). 
 
 Muriatic acid gas is easily dis- FIG. 8. 
 
 tinguished from hydrogen, as well .<^"'lb"-= 
 
 as from common air. It is not in- 
 flammable, and, when exposed to a 
 moist atmosphere, it gives rise to the 
 formation of white clouds (Fig. 8). 
 If a cylinder filled with muriatic 
 acid gas be opened under water, the 
 liquid will rush into the gas, as 
 into a vacuum, filling the vessel 
 completely (Fig. 9). The muriatic acid gas is dissolved in the 
 
LIBEKATION OF HYDEOGEN FROM MURIATIC ACID 
 
 FIG. 9. 
 
 water, and liquid muriatic acid, from which, it was in the first 
 instance expelled, is reproduced. Mu- 
 riatic acid, as well in the form of gas 
 as in a state of solution, affects cer- 
 tain vegetal colours ; changing litmus, 
 for instance, from blue to red, as is 
 shown by dipping into it a litmus- 
 stained strip of paper. 
 
 In order to prove that hydrogen can 
 be procured from muriatic acid by the 
 agency of an alkali-metal, it is only 
 necessary to remove the delivery-tube 
 from the muriatic acid gas apparatus, 
 and to fix in its place, by means of a perforated cork, a tube of 
 difficultly-fusible glass, blown at the middle into a bulb, and 
 containing within the bulb a piece of potassium. These dispo- 
 sitions being made, the muriatic acid gas generated soon reaches 
 the metal, which forthwith becomes covered with a white in- 
 crustation, and if the bulb be now very gently heated by a 
 
 FIG. 10. 
 
 spirit lamp, the potassium fuses, at the same time taking fire, 
 
BY THE ALKALI METALS. 
 
 and burning with a violet light. During this action hydrogen 
 is evolved, and may be ignited at the orifice of the tube (Fig. 10). 
 Sodium produces exactly similar results, but at a much 
 higher temperature. The requisite temperature may, however, 
 be greatly reduced by substituting a solution of sodium in mer- 
 cury for pure sodium. This solution, known amongst chemists 
 as sodium-amalgam, may be obtained by rubbing the two metals 
 together in a mortar, when they unite with powerful evolution 
 of heat, occasionally rising to actual incandescence. Or mercury 
 may be gently heated in a flask, and sodium added in small 
 fragments, which are dissolved with evolution of heat and 
 light. The minute subdivision of the sodium in the amalgam 
 intensifies the reaction between the metal and the muriatic gas, 
 by multiplying their points of contact, so as to bring about the 
 decomposition of the acid at ordinary atmospheric temperature. 
 If it be desired to perform the experiment on a larger scale, 
 the bulb-tube may be replaced by a flask or bottle filled with 
 sodium amalgam. The gas thus obtained is hydrogen; it 
 
 FIG. 11. 
 
 may either be burnt at the orifice of the delivery-tube, or col- 
 lected over water as before (Fig. 11) ; and it is easily recog- 
 nizable by its peculiar properties. 
 
 B 3 
 
10 
 
 AMMONIA ITS DECOMPOSITION 
 
 Pure Ammonia, like muriatic acid, is a gaseous body, of which 
 the so-called liquid ammonia is an aqueous solution. This 
 solution, introduced into a flask and moderately heated, evolves 
 the gas abundantly, as its pungent odour attests. The moist 
 ammonia-gas thus generated requires desiccation ; for which 
 purpose we pass it through an apparatus analogous to that em- 
 ployed for the drying of muriatic acid gas. For reasons, how- 
 ever, to be explained hereafter, the acid-saturated pumice-column 
 with which we previously filled the drying-bottle is, in this case, 
 replaced by a column of fragments of quick-lime ; a substance 
 likewise possessing a strong attraction for water. 
 
 The transparent colourless gas that escapes from the lime- 
 bottle is pure dry ammonia-gas, which, like muriatic acid gas, 
 
 FIG. 12. 
 
 must be collected over mercury (Fig. 12) ; because water 
 absorbs ammonia-gas (Fig. 13) with even greater avidity than 
 we have seen it display for muriatic acid gas. 
 
 From hydrogen, ammonia is readily distinguished by its un- 
 inflammability, by its pungent odour, and by its remarkable 
 solubility in water ; from air it differs by the two properties last 
 mentioned; from muriatic acid gas, by its odour, by its not 
 
BY THE ALKALI METALS. 
 
 11 
 
 Fir,. 13. 
 
 reddening vegetal blues, and by its not fuming when brought 
 into contact with atmospheric air. Ammonia-gas is further 
 characterised by the faculty it pos- 
 sesses of restoring to their primitive 
 colour vegetal blues reddened by the 
 action of acids. This property is 
 most readily displayed by dipping 
 into ammonia-gas, or its aqueous solu- 
 tion, a strip of litmus paper, pre- 
 viously reddened by muriatic acid; 
 when the blue tint instantly reappears. 
 The action of potassium or sodium 
 on ammonia may be exhibited by 
 means of the same apparatus as that 
 employed for the analogous demonstration in the case of muriatic 
 acid gas (Fig. 14). For this purpose also potassium is preferable, 
 
 FIG. H. 
 
 on account of the greater energy with which it acts. So soon as the 
 metal in the glass bulb becomes liquid it is covered with a brownish- 
 green film, and hydrogen begins to escape, as may be proved by 
 lighting it at the mouth of the tube. To collect the hydrogen, 
 
12 
 
 EVOLUTION OF HYDROGEN FROM AMMONIA. 
 
 we introduce some mercury into a glass cylinder, fill it up with 
 water, and then, closing its orifice with a glass plate, plunge 
 it, mouth downward, into a trough of mercury. The orifice of 
 the delivery-tube being brought beneath the inverted vessel 
 (Fig. 15), the gas rises, first through the mercury, then through 
 the water, into the upper part of the cylinder. During its 
 passage through the water, this solvent frees the hydrogen from 
 such intermixed ammonia as it may have retained. The em- 
 ployment of the mercury is a very necessary precaution. Were 
 the delivery-tube allowed to dip directly into water, the portion 
 of ammonia-gas that had escaped decomposition might, from its 
 powerful attraction for water, cause the liquid to rise in the 
 delivery-tube. The water might thus reach the tube containing 
 the heated metal ; whereupon an explosion would inevitably ensue, 
 and shatter the apparatus. 
 
 FIG. 15. 
 
 These methods of obtaining hydrogen, by the action of the 
 alkali metals on hydrochloric acid, water, and ammonia, are cited 
 here, not as the best or easiest processes for procuring that gas, 
 but only as illustrative of the nature of those compounds the 
 
OTHER SOURCES OF HYDROGEN. 
 
 13 
 
 subjects of our immediate study. When the chemist's object is 
 merely to procure hydrogen gas in abundance, he resorts to 
 much easier and less expensive processes. Thus, for example, a 
 plate of zinc immersed in liquid muriatic acid, FIG. ie. 
 
 gives rise to a violent evolution of gas, which 
 may be lighted at the mouth of the vessel (Fig. 
 16), and is thus easily recognized as hydrogen. 
 If the zinc be allowed to act upon the acid 
 in a two-necked bottle, provided with a deli- 
 very-tube, the gas may be collected in inverted 
 cylinders over water (Fig. 17). By this pro- 
 cess large quantities of hydrogen may be 
 easily and economically prepared. Instead of muriatic acid, 
 lastly, dilute sulphuric acid may very conveniently be used as a 
 source of hydrogen gas. But with these processes we are not 
 at present concerned. For our immediate purpose, it is enough 
 to know that hydrogen may be procured from three different 
 substances muriatic acid, water, ammonia ; and that it may be 
 separated from each of these bodies by the intervention of the 
 same metals, potassium and sodium. 
 
 FIG. 17. 
 
 
 It is, however, proper here to observe that hydrogen may also 
 be obtained from muriatic acid, from water, and from ammonia, 
 by the aid of a peculiar force, called Electricity ; which, among 
 many other marvellous properties, possesses that of impressing 
 
14 ELECTRICITY ITS DEVELOPMENT. 
 
 upon bodies certain modifications of the class which we have 
 already defined as chemical. 
 
 It does not fall within our present scope to dilate upon the 
 nature of electricity (a name derived from r/XeK-poi', amber, in 
 which resin this force was first generated by friction) ; still less 
 do our limits permit us to describe the various forms of appara- 
 tus by means of which it is developed. The plan of this course 
 debars us from entering on these topics, which belong to the 
 domain of Physics. Indeed, in true logical order, the study of 
 physics should precede that of chemistry ; and I might, there- 
 fore, fairly assume you to be acquainted with the physical forces 
 and their laws. At all events, it must suffice that we here recall 
 to memory a few particulars which it is essential that we should 
 bear in mind, in order to avoid periphrasis in our future inves- 
 tigations. 
 
 The apparatus employed in developing the electric force which 
 we are about to employ is called a battery ; and the two wires 
 by which the force generated in an electric battery is trans- 
 mitted and applied, are termed the battery poles or electrodes 
 (from r]\KTpov and 6o'c, meaning roads for electricity). In every 
 battery one pole is termed the positive, the other the negative 
 electrode ; and in the form of battery we shall employ (which 
 consists of zinc and carbon couples), the positive pole proceeds 
 from the carbon, and the negative pole from the zinc end of the 
 battery. As these poles have to be dipped in the fluids operated 
 on, which are often of a corrosive character, it is found advisable 
 to arm them at their extremities with some incorrodible substance, 
 such as platinum or carbon. 
 
 At our next meeting we shall proceed to consider the curious 
 and instructive results obtained when the electric force, thus 
 generated and applied, is brought to bear on muriatic acid, water, 
 and ammonia, respectively. 
 
15 
 
 LECTUKE II. 
 
 Electrical investigation of muriatic acid, water, and ammonia action of the 
 electric current on muriatic acid development therefrom of mixed hydro- 
 gen and chlorine separation of chlorine from the mixture distinctive 
 properties of chlorine reproduction of muriatic acid, by synthesis, 
 from hydrogen and chlorine whence its modern name hydrochloric 
 acid action of the electric current on water development therefrom of 
 mixed hydrogen and oxygen separation thereof distinctive properties 
 of oxygen liberation of oxygen from water by chlorine synthetical 
 reproduction of water from hydrogen and oxygen action of the electric 
 current upon ammonia evolution therefrom of mixed hydrogen and 
 nitrogen separation thereof distinctive properties of nitrogen its libera- 
 tion from ammonia by means of chlorine synthesis of ammonia from 
 its elements not yet achieved proofs of the composition of ammonia- 
 matter, simple and compound table of elements. 
 
 WHEN a current of electricity is caused to pass through muriatic 
 acid, by immersing therein the poles or electrodes of a battery 
 (Fig. 18), minute gas bubbles are immediately seen rising from 
 
 FIG. 18. 
 
 their extremities, whilst a peculiar suffocating odour escapes 
 from the fluid. If the experiment be made in a close vessel, 
 these phenomena may be more accurately examined. For this 
 
16 
 
 MURIATIC ACID ITS ELECTROLYSIS; 
 
 purpose the acid to be operated on is placed in a small cylinder, 
 closed by a cork, through which pass the two electrodes, to- 
 gether with a delivery-tube, so that the gas may be collected, 
 as already explained, in a cylinder inverted over the water- 
 trough (Fig. 19). 
 
 FIG. 19. 
 
 On applying a light to the gas thus obtained, it is found to 
 be inflammable, a fact at once suggestive of the presence of 
 hydrogen. But the peculiar suffocating smell already men- 
 tioned does not belong to hydrogen, nor does hydrogen possess 
 the bleaching property which is found to be inherent in the 
 gas escaping from the cylinder. Experimental proof of this 
 bleaching property is readily obtained. For this purpose it 
 suffices to colour the muriatic acid used with a few drops of 
 indigo-solution, which is observed to become rapidly colourless 
 as the gas is evolved; or a strip of litmus paper may be 
 exposed to the current of the gas, which readily bleaches it. It 
 is therefore evident that some gas, as yet unknown to us, is 
 mixed with the escaping hydrogen. 
 
 It now, of course, becomes our object to ascertain the nature 
 of the unknown gas, and for this purpose to obtain it alone. Its 
 isolation may be accomplished by means of a V-shaped glasd 
 tube, with one closed and one open limb; the former being 
 provided with a platinum-wire fused into the glass, through 
 which it passes to terminate, near the bend of the V, in a slip 
 of platinum-foil. Into this tube muriatic acid of I'l spec, 
 grav., coloured with indigo-solution, is introduced, so as to fill 
 
EVOLUTION OP HYDKOGEN THEREBY. 
 
 17 
 
 the whole length of the sealed, and about half the length of 
 the open, limb. For the purpose of decomposing the acid we 
 connect the negative pole of the battery with the wire of the 
 sealed limb, at the same time inserting the positive pole, 
 through the open mouth of the apparatus, into the liquid 
 (Fig. 20). We now observe that gas is almost exclusively 
 
 FIG. 20. 
 
 evolved at the negative pole ; so slight, indeed, is the disen- 
 gagement of gas at the positive pole, that it would scarcely 
 attract attention but for its suffocating odour, and its powerful 
 
 FIG. 21. 
 
 action upon the indigo-coloured solution, which it rapidly 
 bleaches. The gas developed at the negative pole in the sealed 
 
18 
 
 CHLORINE ITS ELECTROLYTIC DISENGAGEMENT ; 
 
 limb has no such bleaching powers, but leaves the blue colour 
 of the liquid unchanged. So soon as a sufficient quantity of gas 
 has been collected in the sealed limb (a result usually obtained 
 in eight or ten minutes), the electric current is interrupted, and 
 the gas is transferred to the open limb, previously filled up with 
 water and closed with the thumb (Fig. 21). It is found to be 
 inflammable, and we recognize it without difficulty as hydrogen. 
 The experiment is now reversed by connecting the positive pole 
 with the sealed, and the negative with the open, limb (Fig. 22) ; 
 
 FIG. 22. 
 
 
 and immediately, as might be expected, hydrogen escapes in 
 abundance from the open mouth of the bent tube, while the 
 liquid becomes decolorized in the closed limb. But this altered 
 disposition of the experiment enables us to acquire some 
 further information regarding the second gas, that, namely, 
 which is evolved in small quantity, but marked by its power- 
 fully irritating odour, and by its bleaching action upon 
 vegetal colours. In the course of ten or fifteen minutes the 
 decolorized liquid in the sealed limb begins to assume a 
 yellowish-green colour, and the evolution of gas (at the outset 
 scarcely perceptible) becomes gradually more and more copious ; 
 so that, in thirty or forty minutes, the greater portion of the 
 tube is filled with a transparent yellowish-green gas. The 
 
ITS ORDINARY PREPARATION. 
 
 19 
 
 battery current is now interrupted, and the gas transferred to 
 the open limb for examination. The approach of a light proves 
 it to be uninflammable : the mouth of the tube is no sooner 
 unclosed than its suffocating odour, already mentioned, be- 
 comes most plainly perceptible, while its powerful bleaching 
 property is shown by its decolorizing, in the very act of trans- 
 ference to the open limb of the tube, the portion of the acid 
 which had retained its blue colour. This peculiar gas has 
 received the name of Chlorine, derived from the Greek word 
 xXwpoc (yellowish-green). 
 
 Chlorine may be obtained from muriatic acid by another 
 process, which has the great advantage of evolving it without 
 giving rise to a simultaneous disengagement of hydrogen. 
 Muriatic acid, when heated with powdered black oxide of man- 
 ganese, in a flask (Fig. 23), yields abundance of chlorine, which 
 
 FIG 23 
 
 may be collected in inverted gas-jars, over tepid water. The 
 manganese process is universally adopted when large quantities 
 of chlorine are required. Hereafter, when we come to study 
 more minutely the nature of this important body, we shall have 
 occasion to dwell on the details of this process : at present it 
 
20 
 
 CHARACTERS OF CHLORINE. 
 
 only claims our attention, in passing, as an easy and economical 
 mode of evolving chlorine in abundance, for the purpose of 
 studying its remarkable properties. 
 
 Chlorine is soluble in about one-third of its volume of cold 
 water : a property which explains its slow and scanty evolution, in 
 the form of visible bubbles, at the outset of the experiment, and 
 the more rapid disengagement of the gas at a later period, when 
 the liquid is saturated therewith. This solvent power of water 
 for chlorine is, however, greatly diminished by heat ; whence the 
 recommendation to collect the gas over tepid water. A lighted 
 taper, when introduced into a vessel containing chlorine, burns 
 with a dim flame, depositing a good deal of soot. Of this circum- 
 stance advantage may be taken to make manifest the high volume- 
 weight of this gas relatively to that of air and, a fortiori, of 
 hydrogen. When a cylinder filled with chlorine is held mouth 
 downward, the colour of the gas rapidly disappears, and its 
 odour is no longer perceptible. If a lighted taper be now 
 introduced into the cylinder, its unimpeded combustion indi- 
 cates that the chlorine has been displaced by air. Again, if 
 a taper be allowed to burn at the bottom of an air-filled vessel 
 
 FIG. 24. 
 
 (Fig. 24), and the mouth of a cylinder filled with chlorine be 
 inclined towards it, as if water were being poured, the flickering 
 
REPRODUCTION OF MURIATIC ACID. 21 
 
 and smoky flame shows that chlorine is falling on the taper. 
 Thus we learn that chlorine is heavier than air, and conse- 
 quently very considerably heavier than hydrogen. Exact ex- 
 periments have proved that, bulk for bulk, chlorine is 35 times 
 heavier than hydrogen ; in other words, that, if the volume- 
 weight of hydrogen be expressed by 1, that of chlorine is re- 
 presented by 35*5. 
 
 From these experiments we have learned that the action of 
 the electric current upon muriatic acid gives rise to the evolu- 
 tion of two essentially different gases ; of which, one, hydrogen, 
 familiar to us from our previous experience, is disengaged at 
 the negative pole, while the other, chlorine, our new acquaint- 
 ance, makes its appearance at the positive pole of the battery. 
 We know, moreover, that each of these gases may be separately 
 evolved from muriatic acid; the hydrogen by sodium, the 
 chlorine by black oxide of manganese ; and we are therefore 
 justified in considering hydrogen and chlorine as constituents of 
 muriatic acid. 
 
 That muriatic acid contains no other than these two consti- 
 tuents remains to be proved by a further experiment. For this 
 purpose it is necessary to obtain a mixture of the two gases in 
 the proportions in which they combine to form muriatic acid. 
 Such a mixture is most readily obtained by the decomposition of 
 muriatic acid itself into its elements ; a decomposition readily 
 effected by the action of the electric current, or, as it is termed, 
 by electrolysis (the terminal, lysis, being derived from the Greek 
 Xvw, I loosen). Let this gaseous mixture FlG 25 
 
 be collected over warm water in a glass- 
 
 stoppered gas-cylinder (Fig. 25), care 
 being taken to allow a considerable quan- 
 tity of the gas to escape before com- 
 mencing the collection. Let the cylinder, 
 when full, be closed, and allowed to stand 
 for several hours in diffuse daylight, and 
 ultimately be exposed to the direct in- 
 fluence of the solar beams. Under this treatment the gas will 
 
22 INFLUENCE OF LIGHT AND HEAT ON THE MIXED GASES. 
 
 be found to have entirely lost its yellowish tint, and to be 
 no longer capable either of taking fire, or of bleaching 
 vegetal colours. In contact with common air the colourless 
 gas forms white clouds ; blue litmus paper exposed to its in- 
 fluence turns red ; and if the cylinder containing it be opened 
 under water the gas is rapidly absorbed. These are the cha- 
 racters of muriatic acid ; and it is evident, therefore, that the 
 hydrogen and chlorine have, in this experiment, reunited to 
 form once more the substance from which they were originally 
 separated. We arrive, of course, at exactly the same result if 
 we make use of hydrogen and chlorine developed from muriatic 
 acid by other means than electricity ; as, for instance, by the 
 action of sodium and of oxide of manganese. 
 
 By whichever of these methods the two gases are obtained, 
 let them be collected in two separate cylinders ; let these be 
 placed one over the other, the lower one upright, the upper one 
 inverted, so that their orifices may meet ; then let the glass 
 plates by which they are closed be withdrawn (Fig. 26), and the 
 
 FIG. 26 
 
 cylinders be shaken, so as to promote the intermingling of the 
 gases ; lastly, let the cylinders be separated, so as to admit of a 
 flame being presented to their orifices (Fig. 27), and then it 
 will be seen that the mixed hydrogen and chlorine, at the touch 
 of fire, instantaneously combine. A sort of hissing explosion 
 
ANALYSIS SYNTHESIS 23 
 
 attends this action, the flame descends into the vessels, and from 
 these dense clouds of muriatic acid escape into the air. 
 
 FIG. 27. 
 
 We have thus endeavoured to determine the nature of muriatic 
 acid by two processes ; firstly, by decomposing it into its con- 
 stituents, secondly, by reproducing it from the constituents 
 previously separated. The former process, that of decomposi- 
 tion, is termed the analytical method (from the Greek araXvw, 
 I unloose) ; the second, that of recomposition, the synthetical 
 method (from the Greek awrtSrjfji, I put together). By the 
 analytical method we have found hydrogen and chlorine to be 
 constituents of muriatic acid ; by the synthetical method we 
 have proved hydrogen and chlorine to be its only constituents, 
 from which circumstance indeed it derives its ordinary chemical 
 appellation, hydrochloric acid. 
 
 We will now apply these two methods, analysis and synthesis, 
 to the better investigation of the nature of water ; of which we 
 as yet only know that, like hydrochloric acid, it evolves hydro- 
 gen when submitted to the action of sodium. 
 
 The experience acquired in the study of hydrochloric acid 
 points out the steps we have to take for the analytical investiga- 
 tion of water. On immersing the platinum poles of the battery 
 in water, to which a little sulphuric acid has been added for 
 the purpose of increasing its conducting power, we obtain at 
 once, in the copious evolution of gas which ensues, abundant 
 proof of the powerful action exerted by the battery upon the 
 
24 
 
 ELECTROLYSIS OF WATER ; 
 
 liquid (Fig. 28). If the current be caused to act upon the 
 water in a close vessel, such as, for instance, the small cylinder 
 
 (Fig. 29) fitted up for the analogous experiment on hydro- 
 chloric acid, we produce and collect a transparent colourless 
 gas, the inflammability of which suggests the presence in it of 
 hydrogen. But the explosive violence with which the gas 
 burns, the sudden flash of the flame to the very bottom of the 
 cylinder, sufficiently indicate that the hydrogen evolved in this 
 process is mixed with some other gas. For the purpose of 
 
 FIG. 29. 
 
 separating the two gases, we resort again to our V-shaped tube, 
 which, having filled with acidulated water, we connect, by the 
 platinum wire at its closed end, with the negative battery pole, 
 at the same time inserting the positive pole in the open limb 
 
GASES EVOLVED THEREBY HYDROGEN OXYGEN. 
 
 25 
 
 (Fig. 30). Streams of gas bubbles appear simultaneously and 
 copiously at both poles, but more profusely at the negative 
 
 FIG. 30. 
 
 pole. Examination of the gas collected in the sealed limb 
 shows that it consists (exactly as in the similar experiment on 
 hydrochloric acid,) of hydrogen (Fig. 31). The experiment is 
 
 FIG. 31. 
 
 now repeated with the electrodes reversed (Fig. 32) (as before), 
 and the hydrogen disengaged at the negative pole allowed to 
 escape into the air. The gas previously evolved at the positive 
 pole in the open limb, and consequently lost, is now collected 
 in the sealed end of the apparatus. 
 
 This gas, like hydrogen, is transparent and colourless; but 
 
 c 
 
26 
 
 CHARACTERS OF OXYGEN. 
 
 it proves on examination to differ essentially from hydrogen. 
 It is not itself inflammable, but it gives intense brilliancy 
 to the combustion of burning bodies, as, for instance, to the 
 
 FIG. 32. 
 
 flame of a wax taper immersed in it (Fig. 33). A wooden splin- 
 ter, retaining but a single ignited spark, when brought into 
 contact with this gas, instantly exhibits a vivid incandescence, 
 
 FIG. 33. 
 
 and in the next moment bursts into flame. This gas, which 
 has received the name of Oxygen, is marked, more strongly than 
 either of the gases which we have hitherto examined, by the 
 characteristic properties of common air. It may, indeed, be inci- 
 
ITS SEPARATION FROM WATER BY CHLORINE. 27 
 
 dentally mentioned here that this gas is a principal constituent 
 of air. 
 
 Oxygen gas is somewhat heavier than atmospheric air. This 
 is readily proved by a simple experiment. Let two cylinders 
 be filled with oxygen, and one of them be placed mouth upward, 
 while the other is suspended mouth downward ; both being un- 
 covered. After the lapse of a few minutes, let the lighted taper, 
 or the faintly-glowing wood splinter, be brought successively 
 to the mouth of each jar to test the quality of its contents. 
 The ensuing sudden and vivid combustion will prove the 
 upright cylinder still to contain oxygen, while the absence of 
 these phenomena will show that the oxygen previously con- 
 tained in the inverted jar has flowed downward through the 
 lighter air, which has ascended to take its place in the cylinder. 
 There is, however, but little difference between the weights of 
 equal volumes of air and oxygen. The relative volume-weight 
 or specific gravity of oxygen, as determined by accurate experi- 
 ment, is represented by 16, that of common air being 14*4, if 
 the volume-weight of hydrogen be taken as the unitary standard 
 of comparison =1. 
 
 There are numerous methods by which oxygen may be pre- 
 pared more conveniently and abundantly than by the action of 
 the electric current on water. For the present, however, we must 
 defer the examination of all these processes save one, which is 
 specially suited to our purpose, as it enables us to extract the 
 oxygen from water without simultaneous disengagement of 
 hydrogen. Whilst studying hydrochloric acid, we had occasion 
 to observe how great an attraction exists between hydrogen and 
 chlorine, and how readily these two gases, by their combina- 
 tion, reproduce hydrochloric acid. Mindful of this behaviour, 
 are we justified in presuming chlorine to be capable, under 
 favourable circumstances, of liberating oxygen from water by 
 withdrawing its hydrogen, and combining therewith to form 
 hydrochloric acid ? We know that, at the ordinary temperature, 
 this action does not occur, for we have collected chlorine over 
 water. Nevertheless, the decomposition in question takes place 
 
 c 2 
 
28 
 
 DECOMPOSITION OF WATER BY CHLORINE. 
 
 readily at high temperatures. This may be shown by means of 
 an arrangement of apparatus, which, though somewhat complex 
 in appearance, is, in principle, simple and easily understood. 
 In the larger flask (Fig. 34), chlorine is evolved from hydrochloric 
 
 FIG. 34. 
 
 ff 
 
 acid by means of black oxide of manganese, and the gas escaping 
 from this flask is caused to bubble through hot water contained 
 in the smaller flask. The chlorine, thus saturated with steam, 
 is then passed through a porcelain tube heated to redness in 
 the furnace. The gas which escapes from the red-hot tube is 
 easily recognized as a mixture of oxygen and hydrochloric acid. 
 To separate these two gases it is only necessary to connect 
 with the furnace-tube a wash-bottle filled with water (or solution 
 of soda), which absorbs and retains the hydrochloric acid, allow- 
 ing the pure oxygen to pass on through the delivery-tube into 
 the inverted cylinder. 
 
 We have thus endeavoured to establish the composition of 
 water by analysis ; that is to say, having on the one hand simul- 
 taneously evolved hydrogen and oxygen from water by means of 
 the electric current, and liberated, on the other hand, the hydro- 
 gen by means of sodium, and the oxygen by means of chlorine, 
 
SYNTHESIS OF WATEE. 
 
 29 
 
 we have analytically found hydrogen and oxygen to be con- 
 stituents of water. 
 
 In order to prove, by the converse or synthetical method, 
 that hydrogen and oxygen are the only constituents of water, 
 we have, as already explained, to reproduce the water from 
 hydrogen and oxygen. 
 
 For this purpose we employ a two-necked glass bottle, having 
 a funnel-tube fixed in one of the necks, and a vitriol-charged 
 desiccating-tube in the other ; a delivery-tube, drawn out to a 
 fine orifice, terminates the desiccator. Hydrochloric acid is 
 introduced into this bottle, and zinc is added (comp. p. 13) ; 
 hydrogen is thus liberated, and, losing its moisture in its pas- 
 sage through the desiccator, issues perfectly dry from the 
 terminal jet. Having allowed the gas to escape for some time, 
 we apply a light to the issuing stream, and introduce the burn- 
 ing jet into a bell-jar filled with dry oxygen gas (Fig. 35). 
 
 We observe that the walls of the vessel become immediately 
 bedewed with a film of fluid, which gradually collects into 
 droplets, and which we readily recognize as water. 
 
 Having thus experimentally investigated the association of 
 
30 
 
 AMMONIA: ITS ELECTROLYSIS 
 
 hydrogen with chlorine and oxygen respectively in hydro- 
 chloric acid and water, we may now pursue the same inquiry 
 with reference to the constitution of ammonia, the third hydro- 
 gen-yielding substance to which our attention has been called. 
 
 FIG. 36. 
 
 Ammonia may, like hydrochloric acid, be decomposed by 
 electrolysis ; and although the action of the electric current is, 
 in this case, somewhat less simple, it is by no means less in- 
 structive. The solution of ammonia to be operated on is 
 adapted for the purpose by the addition of a few drops of sul- 
 
 FIG. 37. 
 
 phuric acid to increase its conducting power. The electrodes 
 are then immersed in the liquid, and an effervescence, occasioned 
 by the evolution of gas, is immediately observed to take place 
 
GASES EVOLVED THEREBY. 
 
 31 
 
 (Fig. 36). On allowing the action to proceed in a close vessel, 
 provided with a delivery-tube (Fig. 37), we collect over water 
 a transparent, colourless, inflammable gas, which we are in- 
 clined to consider as hydrogen. But inferring, from the results 
 obtained in the examination of hydrochloric acid and water, 
 the probability that a second gas is present, we have recourse 
 once more to our V-tube, which admits of the gases evolved 
 at the two poles being separately collected. We commence, as 
 
 FIG. 38. 
 
 before, by connecting the negative pole with the sealed limb of 
 the apparatus (Fig. 38). The gas, which rapidly collects in 
 
 FIG. 39. 
 
 the tube, we unhesitatingly recognize as hydrogen (Fig. 39), 
 and we hasten therefore to repeat the experiment with the poles 
 
32 
 
 NITROGEN GAS; ITS CHARACTERS. 
 
 reversed (Fig. 40). We are at once struck with the compara- 
 tively small quantity of gas evolved at the positive pole, the 
 process having to be continued for half an hour at least before 
 the gas has been collected in sufficient quantity for examination. 
 
 FIG. 40. 
 
 In aspect this gas is transparent and colourless, like hydro- 
 gen, from which, however, the approach of the lighted taper 
 proves it to differ essentially in not being inflammable. Nor is 
 it less different from chlorine and oxygen ; the absence of 
 colour and odour as clearly distinguishing it from the former 
 gas, as its behaviour towards burning bodies does from the 
 
 FIG. 41. 
 
 latter. The flame of a taper, when immersed in this gas 
 (Fig. 41), is immediately extinguished. This body, which 
 
ITS SEPARATION FROM AMMONIA BY CHLORINE. 
 
 33 
 
 chemists term Nitrogen, is marked rather by the absence of 
 salient characteristics than by any active properties of its own. 
 Even its volume-weight is, so to speak, of a neutral average 
 character. While hydrogen is very much lighter, chlorine 
 very much heavier, and even oxygen somewhat heavier than air, 
 we find the volume-weight of nitrogen and air to be almost 
 identical. Compared with hydrogen as a standard unit, the 
 volume-weight of nitrogen is 14, that of air being 14'4, as 
 already stated. This will not, however, be matter of surprise 
 hereafter ; when, studying in its turn the composition of atmo- 
 spheric air, we find nitrogen to be its most abundant constituent. 
 Nitrogen may be liberated from ammonia by a process similar 
 to that which enabled us to separate oxygen from water, 
 namely, by the action of chlorine. Even at common tempera- 
 tures chlorine combines with the hydrogen of ammonia, and sets 
 free its nitrogen. For this purpose we pass a current of chlorine 
 gas through the strongest ammonia-solution of commerce con- 
 tained in a large three-necked bottle (Fig. 42). Powerful action 
 
 FIG. 42. 
 
 4- 
 
 Tr ~ - >i^ .. 
 
 
 
 is at once manifested by the formation of white fumes in the 
 upper part of the bottle, while flashes of light traverse the 
 
 c 3 
 
d4 HYDROGEN AND NITROGEN SOLE CONSTITUENTS OF AMMONIA. 
 
 liquid, which is caused to effervesce by the evolution of gas. 
 The white fumes evolved indicate the formation of a solid 
 body, with which we are not at present concerned, further 
 than to mention that it necessitates the use of the wide connect- 
 ing tubes shown in the figure, as small ones would be speedily 
 choked by the condensation of the fumes. The gas obtained is 
 passed through a wash-bottle, and then collected over water 
 for examination; .which soon proves it to be neither hydrogen 
 nor chlorine nor oxygen, but the same gas (nitrogen) which we 
 obtained by the electrolysis of ammonia. 
 
 By means of the electric current, we have analytically proved 
 hydrogen and nitrogen to be constituents of ammonia. The 
 presence of these two gases in ammonia has been, moreover, 
 demonstrated ; that of hydrogen by the action of sodium, that of 
 nitrogen by the action of chlorine. To proceed in accordance 
 with the plan adopted in the examination of hydrochloric acid 
 and water, we should now by synthesis prove hydrogen and 
 nitrogen to be the only constituents of ammonia. Unfortunately, 
 up to this hour no simple process lias been discovered whereby 
 ammonia can directly be reproduced from hydrogen and nitrogen. 
 We must, therefore, for the present rest contented with the fact 
 that the united weights of hydrogen and nitrogen extracted from 
 ammonia have been found to correspond exactly with the weight 
 of the ammonia operated on a result affording irrefragable 
 proof that these two bodies, and no others, enter into the consti- 
 tution of ammonia. 
 
 The study of hydrochloric acid, water, and ammonia, has 
 acquainted us with a series of facts, the importance of which 
 will not be fully apparent to us till we are further advanced 
 in our inquiry. But we have already obtained some partial 
 insight into the wide field of research they open up ; and a 
 brief retrospective glance may be appropriate here to prepare us 
 for proceeding with our investigation. 
 
 Under the influence of electricity, of heat, and of certain 
 chemical agents, we have seen a small number of well-known 
 
ELEMENTARY BODIES ; THEIR CHARACTERS. 35 
 
 substances pass through a series of most remarkable trans- 
 formations. By appropriate treatment, hydrochloric acid has 
 been found to split up into hydrogen and chlorine, water 
 into hydrogen and oxygen, ammonia, lastly, into hydrogen and 
 nitrogen. 
 
 The ingredients, thus dissevered, we have found ourselves 
 able, in the case of hydrochloric acid and water, directly to 
 recombine, so as to call again into existence the compounds we 
 had previously decomposed; and though, in the case of am- 
 monia, direct synthesis proved to be impossible, chemists have 
 ascertained, by means of the balance, the precise weight of each 
 constituent entering into the composition of a weighed quantity 
 of the compound. In this manner, by a twofold demonstration 
 the most cogent that can be conceived, hydrogen and chlorine, 
 hydrogen and oxygen, hydrogen and nitrogen, have been proved 
 to be the true and only constituents, respectively, of hydrochloric 
 acid, water, and ammonia. 
 
 Having thus succeeded in determining the constituents of 
 hydrochloric acid, water, and ammonia, we are naturally led to 
 inquire into the characters and composition of those constituents 
 themselves. 
 
 Is it in our power to resolve hydrogen, chlorine, oxygen, and 
 nitrogen into simpler forms of matter ? and, if so, what are the 
 methods of analysis by which this result may be achieved ? 
 
 To these questions, which have been experimentally pro- 
 pounded to Nature by many of the most illustrious philosophers, 
 as well of the present as of past generations, but one answer 
 has been obtained, viz., that hydrogen, chlorine, oxygen, and 
 nitrogen, are incapable of decomposition by any means as yet 
 at our disposal. They resist the powerful influences of elec- 
 tricity and of heat, even when raised to the highest attain- 
 able degrees of intensity ; and they issue unchanged from every 
 variety and form of chemical reaction hitherto devised in 
 the hope of resolving them into simpler forms of matter. We 
 are therefore justified in regarding hydrogen, chlorine, oxygen, 
 and nitrogen as indecomposable, or simple bodies, termed 
 
36 NUMBER AND CLASSIFICATION OF THE ELEMENTS. 
 
 ELEMENTS, in contradistinction to compound, decomposable 
 bodies, such as hydrochloric acid, water, and ammonia. 
 
 How many of these bodies, simple and compound, are there ? 
 Of compound bodies, we find a series, numbering many thou- 
 sands, in the so-called inorganic kingdom of nature, which com- 
 prises all the diversified mineral constituents of our earth's 
 crust ; while another series, far more complex in composition, 
 and almost innumerable in multitude, exists in the two great 
 provinces, vegetal and animal, which together make up 
 Nature's organic realm. Yet all this boundless variety of matter 
 springs from only 61 primary bodies, or elements, diversely 
 combined. These appear to be the constituents of the celestial 
 bodies, as well as of our earth. The meteorites which at inter- 
 vals reach our planet from the sky are built up of no other 
 ingredients ; and, as a bold but well-founded induction has, of 
 late years, justified the belief that several of them are also con- 
 stituents of the sun and other fixed stars, we are entitled to 
 regard it as possible that many, if not all, of the remainder may 
 also exist in those bodies. 
 
 These 61 elements are arranged alphabetically in the follow- 
 ing table, in which, it will be observed, they are classified in 
 three groups, by the use of types of three degrees of promi- 
 nence. The first group, distinguished by the largest type, com- 
 prises, besides the four elements with which we have become 
 acquainted in these introductory sketches, some 14 more, which 
 may rank with them as the 18 most widely diffused bodies found 
 at the surface of our globe. This class includes the main con- 
 stituents of the Ocean (oxygen and hydrogen), of the Atmosphere 
 (oxygen and nitrogen), and of the Earth's crust (oxygen in com- 
 bination with silicon, carbon, and the metallic constituents of 
 the earths and alkalis). With these are also classified others 
 which, like bromine and iodine for example, though much less 
 abundant, are equally pervasive. Next in typographical promi- 
 nence are ranged some 23 elements, comparatively rare, but all 
 more or less useful in the arts, and many of them, such as cop- 
 per, tin, zinc. &c. ; familiar to us in every-day life. A third 
 
TABLE OF THE ELEMENTS. 
 
 37 
 
 series, distinguised by the smallest type, comprises about 20 
 elements, which may be termed Nature's chemical rarities- 
 bodies occurring so seldom, and in quantities so minute, as to 
 baffle our endeavours to trace the service which they may render, 
 either in the household of nature or in the arts of life. In this 
 group, three other supposed elementary bodies (Erbium, Ter- 
 bium, and Norium), would find their appropriate places, had 
 not recent researches shown their existence to be so problematic, 
 as to render them quite inadmissible in a list of established 
 elements. 
 
 ALPHABETICAL TABLE OF THE ELEMENTS. 
 
 ALUMINIUM. | IODINE. 
 
 RUBIDIUM. 
 
 ANTIMONY. 
 
 1RIDIUM. 
 
 RUTHENIUM. 
 
 ARSENIC. 
 
 IRON. 
 
 SELENIUM. 
 
 BARIUM. 
 
 LANTHANIUM. 
 
 SILICON. 
 
 BERYLLIUM. 
 
 LEAD. 
 
 SILVER. 
 
 BISMUTH. 
 BO RON. 
 
 BROMINE. 
 
 CADMIUM. 
 
 LITHIUM. 
 
 MAGNESIUM. 
 
 MANGANESE. 
 
 MERCURY. 
 
 SODIUM. 
 
 STRONTIUM. 
 
 SULPHUR. 
 
 CAESIUM. 
 
 MOLYBDENUM. 
 
 TANTALUM. 
 
 CALCIUM. 
 
 NICKEL. 
 
 TELLURIUM. 
 
 CARBON. 
 
 CKBIUM. 
 
 NIOBIUM. 
 
 NITROGEN. 
 
 THALLIUM. 
 THORIUM. 
 TIN. 
 
 CHLORINE. 
 
 OSMIUM. 
 
 TITANIUM. 
 
 CHROMIUM. 
 
 OXYGEN. 
 
 TUNGSTEN. 
 
 COBALT. 
 
 PALLADIUM. 
 
 URANIUM. 
 
 COPPER. 
 
 PHOSPHORUS. 
 
 VANADIUM. 
 
 DIDYMIUM. 
 
 PLATINUM. 
 
 YTTRIUM, 
 
 FLUORINE. 
 
 POTASSIUM. 
 
 ZINC. 
 
 GOLD. 
 
 RHODIUM. 
 
 ZIRCONIUM. 
 
 HYDROGEN. 
 
 
 
 The typography of this table makes it apparent at a glance 
 that only about one-third of the elements are entitled to rank as 
 of primary, and a somewhat similar number as of secondary im- 
 portance. It is to these two classes of elements, of course, that 
 our attention will be principally directed. On the "chemical 
 rarities " we shall bestow but a cursory glance. 
 
 From the manner in which we have arrived at the idea 
 of element, it is obvious that this term must be used with cer- 
 tain restrictions. The bodies enumerated in the above table we 
 
38 POSSIBLE DECOMPOSITION OF BODIES DEEMED ELEMENTARY. 
 
 consider to be elements, because, with the knowledge at our 
 disposal we have not at present the means of decomposing them. 
 Possibly the progress of science may reveal these means to 
 a future generation, so that bodies which for us are elements 
 may cease to be such for our successors. From the age of the 
 four so-called "classical elements" (none of them elements for 
 us), down to comparatively recent times, the annals of science 
 record many of these progressive simplifications ; and we should 
 be presumptuous to doubt the possibility of their recurrence 
 hereafter. 
 
39 
 
 LECTURE III. 
 
 Compound bodies their volumetric constitution and condensation ratios 
 as exemplified in hydrochloric acid in water in ammonia illustrative 
 experiments. Mechanical mixture and chemical combination their dis- 
 tinguishing characteristics experimental illustrations thereof mixtures 
 and combinations of the elements of hydrochloric acid of water 
 constancy of chemical composition marked changes of property attend- 
 ing chemical combination. 
 
 FOR the purpose of gaining additional points of view, we resume 
 the study of the compounds, hydrochloric acid, water, and 
 ammonia, which we have selected to inaugurate our studies ; 
 and we will proceed to examine, in the first place, the 
 volumetric proportions in which the elements, hydrogen, chlo- 
 rine, oxygen, and nitrogen combine to form these compounds. 
 
 In order to ascertain the volume ratio in which hydrogen and 
 chlorine combine to form hydrochloric acid, we must repeat the 
 decomposition of this compound, under conditions calculated to 
 permit the measurement of the hydrogen evolved. 
 
 For this purpose we employ a U-shaped glass tube, about 50 
 centimetres long by 1*5 in diameter, having one sealed and one 
 open limb ; and we fix it upon a convenient stand. Just above 
 the bend of the tube, its open limb has a small outlet tube 
 (blown on at the lamp), and to this is affixed a piece of 
 caoutchouc tubing, with an elastic wire nipper attached, by 
 the action of which the caoutchouc tube is pinched close, 
 but can be readily opened at pleasure. This arrangement, by- 
 the-way, is often a convenient substitute for the ordinary stop- 
 cock ; and it is in such constant use in the laboratory that, to avoid 
 periphrasis in referring to it, we will, if you please, agree to call 
 
VOLUMETRIC PROPORTIONS 
 
 it a nipper-tap. It is shown of full size in Fig. 43, side by side 
 with another little arrangement of like kind, in which the com- 
 
 FIG. 43. 
 
 FIG. 44. 
 
 press! on of the elastic tube is effected by means of a screw 
 (Fig. 44). The use of the nipper-tap, or screw-tap, in the pre- 
 sent case, is to facilitate the introduc- 
 tion of an appropriate volume of the 
 gas to be examined into the apparatus. 
 For this purpose the U-tube is first 
 filled with mercury, and then the 
 nipper-tap is set open, so as to afford 
 a gradual exit to the metal in the open 
 limb. The delivery-tube of an appro-, 
 priate gas-generating apparatus is then 
 passed down the open limb to the bend of the tube, in such 
 manner that the gas bubbles up through the mercury into the 
 sealed limb, from which, of course, the mercury escapes as the 
 gas enters, volume for volume. An appropriate quantity of dry 
 hydrochloric acid gas having been thus introduced, the nipper- 
 tap is closed, and mercury is poured into the apparatus, until 
 it stands at the same level in both limbs. The space occu- 
 pied in the tube by the gas is then marked in any convenient 
 way; preferably by a caoutchouc ring slipped over the tube 
 (Fig. 45). 
 
 That portion of the open limb which is unoccupied by 
 mercury is then filled with sodium-amalgam (Comp. p. 9), and 
 the orifice of the tube is closed, either by the thumb or by a 
 glass stopper. The gas may now, by inclining the tube 
 
OF THE CONSTITUENTS OF HYDROCHLORIC ACID GAS. 
 
 41 
 
 FIG. 45. 
 
 adroitly, be easily transferred from the sealed to the stoppered 
 limb ; traversing of course, in its passage, the column of sodium 
 amalgam, and being thereby de- 
 composed. To insure complete 
 decomposition, the apparatus 
 should be once or twice shaken, 
 so as to bring every portion of 
 the gas into thorough contact 
 with the amalgam ; after which, 
 by reversing the previous in-* 
 clination of the tube, the gas 
 may be re-transferred to the 
 sealed limb of the apparatus. 
 On removing the stopper or 
 thumb from the mouth of the 
 
 open limb, the mercury falls a little therein, and may be further 
 lowered by opening the nipper-tap. As soon as the mercury 
 stands at a uniform level in the two limbs, the gas is found 
 reduced to exactly half its ' 
 original volume (Fig. 46). 
 The gas which remains, it need 
 scarcely be remarked, is hydro- 
 gen, readily recognizable by 
 its inflammability. 
 
 This experiment shows us 
 that any given bulk of hydro- 
 chloric acid contains half that 
 bulk of hydrogen. It only, 
 therefore, remains to deter- 
 mine with what proportion, 
 by volume, of chlorine this 
 bulk of hydrogen is combined in hydrochloric acid, 
 
 This we learn from a second experiment. We again submit 
 hydrochloric acid to electrolysis in the apparatus, and with the 
 results already described (comp. p. 15). Hydrogen is copiously 
 evolved at the negative pole, while chlorine is simultaneously 
 
 FIG. 46. 
 
 
42 COLLECTION OF THE GASES PRODUCED 
 
 liberated at the positive pole. At starting, however, the chlo- 
 rine is almost entirely absorbed by solution in the surrounding 
 liquid ; nor is it till this is saturated that the chlorine begins to 
 be manifested in a stream of bubbles, like those which, from the 
 first, mark the escape of hydrogen at the opposite pole. At 
 this stage of the process, the delivery-tube of the apparatus is 
 attached, by means of a caoutchouc connector, to a glass tube, of 
 about 40 or 50 centimetres long by 1*5 centimetre in diameter, 
 drawn out, before the lamp, to a fine point at each end. This 
 tube is then filled with the mixture of hydrogen and chlorine 
 evolved by the electrolysis of hydrochloric acid. In order to 
 expel every trace of air it is necessary that the mixed gases 
 should be suffered to traverse the tube for a considerable time, 
 and, as chlorine is a very noxious gas, it is necessary to take 
 means for preventing its escape into the air. For this purpose, 
 the free end of the tube is connected with the lower part of an 
 upright cylinder containing coke, moistened with an alkaline 
 liquid capable of absorbing the chlorine. Through this appa- 
 ratus, which is termed a coke column, or coke tower, the gas 
 streams up, parting with its chlorine as it ascends ; so that 
 only the hydrogen, which is innocuous, escapes. After the 
 lapse of one or two hours the operation may be considered com- 
 plete. The tube being now detached, its fine-drawn ends are 
 immediately sealed. 
 
 This manipulation, by-the-way, requires considerable precau- 
 tion, when performed upon explosive mixtures such as that with 
 which we have now to deal. The object to be attained is the 
 prevention of contact between the flame and the explosive gas 
 during the sealing of the tube. Such contact would inevitably 
 take place were the flame applied to seal the tube at its extreme 
 tip or orifice. To avoid such contact in other words, to keep a 
 layer of glass interposed between the gas and the flame this 
 latter should be directed, not upon the orifice, but on the long 
 drawn neck between the orifice and the body of the tube. This 
 neck, being slender, soon softens ; and its sides, running together, 
 obliterate the channel, and so seal the tube. The excess of neck 
 
BY THE ELECTROLYSIS OF HYDROCHLORIC ACID. 43 
 
 beyond the point thus sealed may then, while the glass is still 
 soft, be detached, so as to leave a neatly-formed extremity. 
 Thus conducted, on a neck sufficiently drawn out and attenuated, 
 the sealing operation is tolerably safe ; though, to guard against 
 injury from accidental explosion, through the overheating or 
 cracking the glass during the process, it is prudent to envelop 
 the body of the tube in a cloth. 
 
 The tube having been sealed at each end, its gaseous contents 
 have next to be examined. For this purpose it is requisite to 
 bring the mixed gases into contact with a fluid capable of absorb- 
 ing the chlorine but not the hydrogen. Water answers this 
 purpose, and if a little soda be mixed with it the absorptive 
 power is increased. Again, the addition of a vegetal colour of 
 an infusion of logwood, for instance to tint the soda-solution 
 employed, is useful as a means of evincing the presence of chlo- 
 rine by exhibiting its bleaching action on the colour, so soon as 
 it comes into contact therewith. By plunging the sealed finely- 
 drawn extremity of the tube into a solution so prepared, and then 
 breaking it off, the desired contact is effected, absorption begins, 
 and the liquid is seen slowly rising into the tube to occupy the 
 space vacated by the absorbed chlorine. This absorption goes 
 on very slowly, however, because of the extreme minuteness of 
 the surface of fluid exposed to the gas in the finely-drawn 
 tube represented by the broken extremity. A great acceleration 
 would evidently be obtained if the surface of contact could be 
 extended ; if, for example, we could wet the whole in- FI<J 47 
 terior surface of the tube with the absorptive liquor. 
 
 We have here a contrivance for accomplishing this 
 object (Fig. 47). 
 
 It consists of a caoutchouc connector tightly fitted to 
 the end of the tube, so as to cover and enclose its fine- 
 drawn sealed neck. This connector is provided with a 
 small glass funnel, through which it can be filled with a 
 tinted solution of soda , and has also a stop-cock, by 
 which, when so filled, it can be closed. These ar- 
 rangements being made, the* fine-drawn neck becomes im- 
 
44 
 
 ABSORPTION OF THE CHLORINE EVOLVED. 
 
 mersed in the solution ; so that, on breaking it (Fig. 48), 
 which the flexibility of the connector allows to be easily 
 done, the solution finds its way through the orifice into 
 the interior of the apparatus. By suitably inclining this, 
 
 Fro. 48. 
 
 the solution may be caused to extend in a film over its in- 
 terior, so as to expose a widely-spread surface to the gaseous 
 mixture. The absorption of the chlorine is thus made to pro- 
 ceed with greatly increased rapidity, as is evinced by the speedy 
 ascent of a small volume of the liquor into the wide part or body 
 of the tube. This accomplished, that end of the tube which 
 is armed with the funnel may be plunged under water, the 
 flexible connector withdrawn, and the experiment continued in 
 the ordinary way, by allowing the absorption to proceed, and the 
 column of liquor to ascend in the tube till all the chlorine is 
 dissolved. This is known to have taken place by the liquor's 
 ceasing to rise in the tube. 
 
 The tube is now to be more deeply immersed in the liquor 
 (the receptacle for which should be a tall glass cylinder, to 
 facilitate this part of the manipulation), until the water-level 
 
' 
 
 CONSTITUENTS OF HYDROCHLORIC ACID. 45 
 
 **l 
 
 within and without the tube is brought into coincidence. It is 
 found that the tube is just half filled with liquor in other 
 words, that just half its gaseous contents have been absorbed. 
 That the absorbed gas is chlorine is readily proved by the 
 bleaching effect exerted by it on the logwood solution. 
 
 The nature of the residuary gas is as readily demon- 
 strated by immersing the tube more deeply in the surround- 
 ing liquor, then breaking off its upper finely-drawn point 
 and applying to the jet of gas, thus forced out by water 
 pressure, a lighted taper ; when it immediately takes fire, and 
 burns with the characteristic pale-blue flame of hydrogen. 
 
 These phenomena furnish a simple and satisfactory reply to 
 the question left unanswered by our previous experiment. 
 
 The action of sodium upon hydrochloric acid taught us that 
 2 volumes of hydrochloric acid contain 1 volume of hydrogen ; 
 the electrolysis of hydrochloric acid proves that, to form this 
 compound, 1 volume of hydrogen combines with 1 volume of 
 chlorine. 
 
 The two experiments, taken together, supply us with the 
 exact points of information which our previous investigation of 
 hydrochloric acid left deficient ; so that, summing up our 
 previous and present results, we now possess a complete and irre- 
 fragable demonstration, first, that hydrochloric acid is composed 
 of hydrogen and chlorine; secondly, that these two elements 
 are its sole constituents ; thirdly, that they are united in equal 
 volumes to form it ; and lastly, that, in so uniting, they undergo 
 no condensation, but produce a volume of compound gas, equal to 
 the sum of the volumes of its elementary constituents. 
 
 This last-mentioned fact the union of hydrogen and chlorine 
 without contraction or expansion may be illustrated by another, 
 and an equally conclusive, experiment. While the electrolytic 
 apparatus, used in the experiment just made, is still evolving 
 hydrogen and chlorine in the proportion in which the two gases 
 exist in hydrochloric acid, we may replace the wide glass tube, 
 previously used, by another tube of equal length, but of stouter 
 glass, and of smaller bore ; half a centimetre being a convenient 
 
46 SYNTHESIS OF HYDROCHLORIC ACID ; 
 
 diameter. The two ends of this tube are, like those of the tube 
 used in the previous experiment, drawn out into very fine necks. 
 So soon as .the tube is thoroughly purged of air, and exclusively 
 filled with the gaseous constituents of hydrochloric acid, its 
 fine necks are sealed by the blow-pipe jet, and its contents are 
 exposed to the action of light for the purpose of inducing 
 the combination of the mixed gases. 
 
 This curious effect may be obtained either by natural or 
 artificial light. The direct rays of the sun produce instan- 
 taneous combination. But as such rays are not at our com- 
 mand in all seasons and at all places, as, for instance, during 
 the earlier weeks of a London winter, it is convenient to know 
 of an artificial light sufficiently intense to bring about the same 
 effect. Such a light is that of the blue flame evolved by the 
 combustion of two substances called nitric oxide and bisul- 
 phide of carbon ; the properties of which will claim our atten- 
 tion hereafter. At present we need only note the simple 
 manipulation needful for generating this light. 
 
 For this purpose some 8 or 10 cubic centimetres of the fluid 
 called bisulphide of carbon are introduced into a tall glass 
 cylinder filled with nitric oxide gas. This is most conveniently 
 accomplished by means of a very thin bulb of glass, blown to 
 the required size, filled with the bisulphide, and then sealed 
 at the lamp. The glass cover of the vessel, already filled with 
 nitric oxide gas, is drawn aside, the bulb dropped in, and the 
 cover quickly replaced. Contact of atmospheric air is thus almost 
 entirely obviated. The vessel is then jolted, so as to break the 
 glass bulb, and the desired mixture of gas and vapour is at once 
 obtained. A match is now applied to the opened mouth of 
 the cylinder, when the mixture within takes fire, and burns 
 with a brilliant, intensely blue flame, which descends into the 
 vessel. The radiance of this light instantaneously induces the 
 combination of hydrogen and chlorine ; the effect being indi- 
 cated by a flash of light, accompanied by a slight clicking 
 sound, and followed immediately by the disappearance of the 
 greenish colour of the mixture. 
 
APPAKATUS FOR EFFECTING IT. 
 
 47 
 
 The figure (49) shows the disposition of the apparatus. To 
 the left is the glass cylinder in which the light is generated : to 
 the right are placed the mixed gases to be acted on (two tubes, 
 instead of one being filled therewith, and employed in the expe- 
 riment, to afford a double chance of success this being an 
 illustration which occasionally fails from causes not perfectly 
 ascertained. 
 
 FIG. 49. 
 
 For the purpose of examining the product, one point of the 
 tube is broken under mercury ; when the first indication is 
 immediately obtained. It is observed that neither does gas 
 escape from, nor mercury penetrate into, the tube ; from which 
 it is clear that the combination of the gases has taken place 
 without either contraction or expansion of their volume. 
 
 The next indication is obtained by pouring water on the mer- 
 
48 PKECAUTIONS AGAINST EXPLOSION. 
 
 cury, and raising the tube so that its orifice, instead of plunging 
 into mercury, may open into water. A striking effect is now 
 produced. The state of equilibrium which obtained so long as 
 the tube opened into mercury instantly ceases. The water no 
 sooner comes into contact with the gas than this latter is dis- 
 solved ; and so rapid is the absorption of the gas by the water 
 that this rises in the tube, filling it almost instantaneously. The 
 solution thus obtained is readily shown, by the application of 
 appropriate tests as before, to be dilute hydrochloric acid ; and 
 these results confirm our previous experimental proofs that 
 hydrochloric acid is formed by the union of hydrogen and 
 chlorine gases, in equal volumes, without condensation. 
 
 In performing this experiment, whether sunlight or the light 
 of bisulphide of carbon be employed, some manipulatory pre- 
 cautions are necessary to protect the operator from possible 
 injury. The mixed gases in the act of combination evolve so 
 much heat, and are so powerfully dilated thereby, as sometimes 
 to burst the tube containing them. The experimentalist should, 
 therefore, not omit to protect himself by a screen, for which 
 purpose a sheet of stout plate-glass may be conveniently em- 
 ployed. Thus, even should the tube explode, the dangerous 
 scattering of its fragments is prevented. It is, however, only 
 rarely that the body of the tube is shattered : in most cases the 
 fracture is confined to one or other of the sealed points. To 
 avoid the loss of the experiment by an accident of this kind, the 
 upper point of the tube may be strengthened by imbedding it in 
 sealing-wax, which may be most conveniently applied by fusing a 
 little in a small piece of glass tube (see Fig. 49), sealed at one 
 end, and plunging the point to be protected into the fused mass, 
 which is then allowed to cool and harden. As for the lower point, 
 escape of gas from this, in case of rupture, is readily obviated by 
 keeping it plunged into a trough-cylinder filled with mercury. 
 
 The ratio in which hydrogen and oxygen are associated in 
 water is most conveniently established by the electrolysis of 
 this compound. 
 
PROPORTIONS OF HYDROGEN AND OXYGEN IN WATER. 
 
 49 
 
 We have already availed ourselves of the action of the 
 electric current for the purpose of ascertaining the nature of 
 the constituents of water ; and we learned incidentally that the 
 hydrogen was evolved in larger measure than the oxygen. An 
 appropriate modification of the apparatus then used enables us 
 to determine with precision the ratio in which they are evolved. 
 
 For this purpose, two glass tubes of equal diameter, each sealed 
 at one end and open at the other, are filled with acidulated 
 water, and suspended, mouth downward, over a basin also filled 
 with acidulated water, below the level of which the mouths of 
 the two tubes dip (Fig. 50). The conducting-wires, or poles of 
 
 FIG. 50. 
 
 the battery, duly insulated by a gutta-percha covering (except 
 at their points), are immersed in the fluid of the basin, and their 
 points are disposed one below each tube. Each point is armed 
 with a small strip of platinum-foil. 
 
 The battery is now put into action, and its current, trans- 
 mitted by the platinum poles through the water which lies 
 between them, decomposes this into its two constituent gases, 
 which appear, as before, ascending in bubble-streams, one 
 from the positive, the other from the negative pole : this latter 
 stream being (also as before) manifestly the more abundant of 
 the two. 
 
50 
 
 VOLUME-RATIO OF HYDROGEN TO OXYGEN IN WATER. 
 
 As the two gases collect in the respective tubes, it becomes 
 obvious that exactly two volumes of the more abundant are 
 evolved for every one of the less abundant constituent. Now, 
 we know from former experiments that the more abundant 
 gas (that disengaged at the negative pole) is hydrogen, whilst 
 the less abundant product (that liberated at the positive pole) 
 is oxygen. This we can, of course, again verify as before, by 
 removing the tubes, and applying to their contents the tests 
 with which we are already acquainted. It is thus proved that 
 in water two volumes of hydrogen are united with one volume of 
 oxygen. 
 
 FIG. 51. 
 
 Fig. 51 shows an improved form of apparatus for demon- 
 strating this important fact. Instead of the two sealed tubes 
 separately inverted in a basin, as just described, we have here 
 
CONDENSATION OF THE CONSTITUENTS OF WATER-GAS. 51 
 
 a three-branched tube, with one long and two short limbs. 
 The long limb acts as a water-reservoir, instead of the basin ; 
 the two short limbs, which are fitted with stop-cocks, or nipper- 
 taps, above, and which communicate freely with the long limb 
 below, contain, intermediately, each a platinum electrode. 
 When this apparatus is charged with water, and tne electric 
 current is transmitted through the electrodes, the two gaseous 
 products are evolved, each at its appropriate pole, so as to 
 accumulate separately, each in the corresponding limb. As this 
 accumulation proceeds, the water in the short limbs is forced 
 downward out of these, so as to rise in the long limb, forming 
 a column, the weight of which serves, in the sequel, to expel the 
 gas from each short limb through the corresponding stop-cock, 
 when this is opened for the purpose of testing the nature of the 
 gas obtained. 
 
 It remains to be investigated whether hydrogen and oxygen, 
 during their conversion into water, undergo any change of 
 volume ; or, like hydrogen and chlorine, combine to produce a 
 measure of compound gas, exactly equal to the joint measure of 
 the constituents. 
 
 To determine this point it is necessary to compare the volume 
 of the elementary water-constituents with that of the water 
 formed, at a temperature high enough to maintain the latter 
 in the purely gaseous condition in which engineers term it dry 
 steam. FIG. 52. 
 
 The experiment is made in a U tube similar to 
 
 that used in analyzing hydrochloric acid (p. 41). 
 The closed limb of the tube is fitted, at a point 
 near its sealed extremity, with two platinum wires, 
 welded to the glass, which they traverse, and sepa- 
 rated at their inner ends by a distance of only 
 two millimetres ; while their outer ends are formed 
 into loops for the attachment of appropriate bat- 
 tery-conducting wires. 
 
 This arrangement is shown, upon an enlarged 
 scale, in the figure (52), in which two slightly-modified forms 
 
 D 2 
 
 
52 
 
 INDUCTION COIL 
 
 of it are placed side by side ; the only difference being that in 
 one the platinum points are bent slightly up within the tube, to 
 facilitate the cleansing of its interior. 
 
 The general disposition of the apparatus is depicted in Fig. 
 53, which shows the electric arrangements to the left ; and the 
 U tube, wrth its adjuncts, to the right. The so-called " induc- 
 tion coil," interposed in the electric circuit causes the cur- 
 rent to pass in the form of a stream of sparks between the 
 wires within the U tube. Into the sealed limb of this tube, 
 which is filled with mercury, we admit a column, about 25 
 or 30 centimetres high, of a mixture of hydrogen and oxy- 
 gen in the proportions in which they form water. This 
 mixture may of course be obtained by adding two volumes 
 of hydrogen to one of oxygen ; it is, however, prepared much 
 more readily in due volumetric proportion, and in a state of 
 perfect purity, by the electrolysis of water, as already described. 
 The gas-filled limb of the U tube is surrounded by a high 
 glass cylinder, the lower mouth of which is fastened around it 
 
 FIG. 53. 
 
 by means of a perforated cork, whilst its upper mouth (likewise 
 closed by a cork) rises about five centimetres above the sealed 
 extremity. The annular space thus formed communicates, at 
 its upper end, by means of a bent glass tube and a perforated 
 cork, with a flask which contains a liquid having a boiling point 
 considerably above that of water : amylic alcohol, which boils at 
 132 C., is well adapted for the experiment. On protracted ebul- 
 lition, the vapour descends from the flask into the annular space, 
 which rapidly acquires a uniform temperature of 132. To 
 
AND SPARK- STREAM APPARATUS. 
 
 53 
 
 prevent the powerfully-odoriferous vapours from escaping into 
 the atmosphere, the lower extremity of the glass cylinder is con- 
 nected with an appropriate vapour-condenser, such as a glass 
 tube kept cool by a current of water. Under the influence of 
 heat, the column of mixed hydrogen and oxygen in the tube 
 expands, and its height is marked by any suitable means ; prefer- 
 ably by slipping a caoutchouc ring over the outer glass cylinder. 
 Care must of course be taken, before doing this, to bring the 
 mercury to a uniform level in both limbs of the U tube, either by 
 adding or withdrawing a proportion of the metal as required. A 
 little more mercury is then poured into the open limb, which is, 
 lastly, closed by a well-fitting cork. Between this cork and the 
 
 Fm. 53. 
 
 mercury intervenes a column of air, some eight or ten centi- 
 metres in length, and capable of yielding to pressure, like a 
 spring. It now only remains to inflame the gaseous mixture by 
 causing the current of the induction-coil to leap, in the form of 
 a spark, between the platinum points. The gases combine with 
 an explosion ; which is, however, much mitigated in violence by 
 the elastic action of the above-mentioned air column. At the 
 high temperature employed (132), the water formed retains the 
 gaseous condition. On removing the cork, and allowing the 
 
54 VOLUME-RATIO OF HYDROGEN AND NITROGEN IN AMMONIA. 
 
 mercury to flow through the nipper-tap, until it is level in both 
 limbs of the U tube, it becomes obvious that the original mea- 
 sure of mixed gases is diminished by one-third ; the residuary 
 two-thirds are water-gas, or dry steam, which condenses into 
 liquid water so soon as the tube is allowed to cool. 
 
 Thus, therefore, it stands experimentally demonstrated, first, 
 that hydrogen and oxygen undergo condensation in combining 
 to form water ; and, secondly, that the volume of the water-gas 
 produced holds an extremely simple ratio to the volume of its 
 constituent gases ; two volumes of hydrogen and one volume 
 of oxygen condensing, by their union, into two volumes of 
 water-gas. 
 
 The method of ascertaining the volume-ratio in which hy- 
 drogen and nitrogen combine to form ammonia is less simple 
 than that which suffices for the corresponding study of hydro- 
 chloric acid and water. 
 
 For this purpose chlorine is employed, as before, to withdraw 
 hydrogen from ammonia, and set free nitrogen ; and besides 
 this, means are adopted to determine with accuracy the volume 
 of nitrogen thus separated from a known quantity of ammonia. 
 A glass tube for holding chlorine, and a globe for receiving 
 FIG. 54. solution of ammonia, and admitting it, drop by drop, 
 to the chlorine, constitute the requisite apparatus. 
 The glass tube is from 1 to 1-5 metre long, sealed 
 at one end, open at the other, and marked oft, by 
 elastic caoutchouc rings slipped over it and clipping 
 it firmly, into three equal portions. The globe has 
 a stoppered aperture above, and a dropping tube 
 (Fig. 54) drawn out to a narrow orifice below. This 
 tube is fitted with a stop-cock, and passes through a 
 perforated cork, by means of which it can be tightly 
 fixed into the open mouth of the chlorine-tube. 
 
 The apparatus is thus employed. The long chlo- 
 rine-tube having been filled with cold water and inverted over 
 a pneumatic trough, with its mouth immersed below the water- 
 
ITS DETERMINATION BY MEANS OF CHLORINE. 
 
 OO 
 
 \ 
 
 level, is filled with chlorine gas in the usual way. When full, 
 it is still allowed to stand for about fifteen minutes over the 
 chlorine delivery-tube, that its interior surface may be quite 
 freed from the chlorine- saturated water that else would remain 
 adherent thereto. The globe, meanwhile, is filled with a strong 
 solution of ammonia, and its stop-cock is turned so that its 
 dropping-tube also may be filled to its very tip with this solu- 
 tion. The globe is then stoppered, and the cock closed again ; 
 after which it is ready for connection with the chlorine-tube. 
 To effect this connection without admission of air into the 
 chlorine-tube requires some little care and dexterity. The globe 
 has to be immersed in the pneumatic trough, with its dropping- 
 tube upward, and in this position to be brought beneath the 
 mouth of the chlorine-tube, into FIG. 55. 
 
 which the globe-tube is inserted, 
 and fixed firmly by means of the 
 cork which it carries. In effect- 
 ing this junction, great care must 
 be taken not to introduce any 
 water from the trough into the 
 chlorine-tube. This tube, with 
 its ammonia-globe joined to it, 
 may now be removed from the 
 trough, and supported in a ver- 
 tical position, with the globe 
 surmounting it. A single drop of 
 the ammonia-solution is now suf- 
 fered to fall from the globe into 
 the chlorine-tube, the stop- cock 
 being opened for a moment for 
 this purpose (Fig. 55). The en- 
 trance of this drop into the atmo- 
 sphere of chlorine is marked by 
 a small, lambent, yellowish-green 
 flame at the drawn-out point of the dropping-tube. Drop by drop, 
 at intervals of a few seconds, the ammonia-solution is allowed to 
 fall into the chlorine-tube, the ammonia of each drop being at 
 
56 STAGES OF THE OPERATION. 
 
 the instant of its contact with the chlorine converted, with a flash 
 of light and the formation of a dense white cloud, into hydro- 
 chloric acid and nitrogen. The addition of ammonia must be 
 continued till the whole of the chlorine present is supplied with 
 hydrogen at the expense of ammonia. To insure this the 
 ammoniacal solution is added in excess, a column of three or 
 four centimetres being abundantly sufficient. The result is that 
 the hydrochloric acid formed combines with the excess of 
 ammonia to form a compound, of which we shall have to 
 speak hereafter ; but of which it is enough to say here that it 
 makes its appearance as a white deposit, lining the interior of 
 the chlorine-tube. This deposit, being soluble, is readily washed 
 down and dissolved by agitating the liquor in the tube, which 
 now contains the whole of the nitrogen separated, except a little 
 which remains dissolved in the liquor. This small quantity 
 of dissolved nitrogen is easily expelled from the liquor by heat. 
 
 We are now sure of two points ; viz., that the whole of the 
 chlorine has been converted into hydrochloric acid at the ex- 
 pense of the ammonia ; and, secondly, that we possess within 
 our tube the whole of the nitrogen thus set free. 
 
 It becomes our next object to withdraw the excess of ammonia. 
 For this purpose dilute sulphuric acid, which fixes ammonia, 
 is introduced by means of the globe previously employed to 
 admit ammonia. 
 
 The nitrogen, being thus freed from all intermixed gaseous 
 bodies, has only now to be brought to mean atmospheric tempera- 
 ture and pressure in order that it may be ready for measurement. 
 
 The temperature, which had been raised by the application 
 of heat to the liquor to expel the dissolved nitrogen therefrom, 
 is readily brought back to the mean by plunging the tube into 
 cold water. To equalize the pressure within and without the 
 tube, the bent syphon tube (Fig. 56) is employed. One end of 
 this communicates with the interior of the tube, while the other 
 plunges beneath the surface of water subject to atmospheric pres- 
 sure. That this pressure exceeds that of the gas in the tube is 
 at once seen by the flow of the water through the syphon into 
 the tube. As the water-level in the tube rises, the nitrogen? 
 
CONDENSATION KATIO OF THE ELEMENTS OF AMMONIA. 57 
 
 previously expanded, gradually approaches its normal volume, 
 which it exactly attains when the flow ceases, showing the 
 pressure within and without to be in equilibrium. Both tem- 
 perature and pressure being now 
 at the mean, all the requisite 
 conditions are fulfilled for obtain- 
 ing an exact knowledge of the true 
 volume of nitrogen ; and this, on 
 inspection, is found exactly to fill 
 one of the three divisions marked 
 off at the outset on our tube. 
 
 Now, bearing in mind that we 
 started with the three divisions 
 full of chlorine, and that we have 
 saturated this chlorine with hy- 
 drogen supplied by the ammo- 
 nia ; bearing in mind, moreover, 
 that hydrogen combines with 
 chlorine, bulk for bulk ; it is evi- 
 dent that the one measure of nitro- 
 gen which remains in the tube has 
 resulted from the decomposition 
 of a quantity of ammonia contain- 
 ing three measures of hydrogen. 
 
 It is, therefore, clearly proved 
 by this experiment that ammonia 
 is formed by the union of three 
 volumes of hydrogen with one volume of nitrogen. 
 
 Thus much determined, it remains to ascertain the condensa- 
 tion undergone by these elements in combining to form ammonia. 
 
 Of this we may obtain ocular demonstration by a very simple 
 experiment. We cannot, indeed, employ in the case of ammo- 
 nia the method adopted to determine this ratio in the cases 
 of hydrochloric acid and water ; we cannot, that is to say, mix 
 hydrogen and nitrogen in due volumetric proportion to form 
 ammonia, and then cause them to unite, and measure the space 
 
 D 3 
 
58 
 
 DILATATION OP AMMONIA-GAS WHEN DECOMPOSED. 
 
 occupied by the product. The direct synthesis of ammonia has 
 never yet been accomplished ; so that we must fall back upon ana- 
 lysis to furnish us with this demonstration. We must split up a 
 measured quantity of ammonia into its constituents, and compare 
 the space occupied by the ammonia before treatment with the 
 space filled by its separated constituents. This we are enabled 
 to do very easily, by availing ourselves of the tendency of am- 
 monia to break up into its elements under the influence of a 
 moderate heat. 
 
 For this purpose we need our often-employed U tube as a 
 receptacle, and the spark-stream supplied by the electric current 
 in traversing our induction-coil as a source of heat (Fig. 57). 
 
 The sealed limb of the U tube is filled to about one-third of 
 its height with dry ammonia, over mercury, and the height of the 
 column of gas is accurately measured ; care having been taken, 
 as usual, to bring the mercury in each limb of the tube to a 
 uniform level. The spark-stream is now set flowing between 
 the platinum points, and the volume of the gas on which 
 it acts is immediately observed to increase. This dilatation 
 continues for some five or ten minutes (according to the 
 quantity of ammonia under treatment) ; and, when it ceases, the 
 level of the mercury (disturbed, of course, by the expansion of 
 
 FIG. 57. 
 
 the gas) is readjusted to perfect uniformity in both limbs of the 
 tube ; when it is immediately perceived that the original volume 
 of gas has become doubled. If a little of the gas (previously so 
 pungent) be allowed to escape from the tube, by a stop- 
 cock provided for that purpose, it is found to have become 
 
SUMMARY OF CONDENSATION RATIOS. 
 
 59 
 
 FIG. 57. 
 
 inodorous; while the presence of hydrogen 7S indicated by 
 its inflammation on the approach of a light. 
 
 This experiment proves that hydrogen and nitrogen, as com- 
 bined in ammonia, occupy only half the space they fill in their 
 free state ; or, in other words, that 4 volumes of the mixed 
 gaseous constituents of ammonia, composed, as we have already 
 ascertained, of 3 volumes of hydrogen and 1 volume of nitro- 
 gen, condense, during their combination, to form 2 volumes of 
 ammonia. 
 
 The result of these inquiries into the composition of hydro- 
 chloric acid, water, and ammonia may be summed up as fol- 
 lows : 
 
 1 vol. of hydrogen -f 1 vol. of chlorine = 2 vols. of hydrochloric 
 
 acid. 
 
 2 vols. of +1 vol. of oxygen = 2 vols. of water gas. 
 
 3 vols. of -}- 1 vol. of nitrogen = 2 vols. of ammonia. 
 
 Thus it is evident that hydro' 
 chloric acid, water, and ammo- 
 nia, not only differ as to the 
 volume-ratio of their element- 
 ary constituents, but also as 
 to the ratios of the spaces oc- 
 cupied by these before and after 
 their combination to form che- 
 mical compounds. These ra- 
 tios rise from unity, or ^, in 
 the case of hydrochloric acid, 
 through in the case of water- 
 gas, to i in the case of am- 
 monia ; the condensation increasing, in these cases, pari passu 
 with the complexity of the chemical compound. 
 
 Notwithstanding these remarkable differences in their chemi- 
 cal construction, the three compounds, hydrochloric acid, water, 
 and ammonia, illustrate several important general laws, which 
 
60 IMMUTABLE PROPORTIONS OF CHEMICAL COMPOUNDS. 
 
 deserve our particular attention. Among the most strik- 
 ing of these are: 1, the immutability of the proportions in 
 which the elements are in each case associated ; and, 2, the 
 strongly-marked changes of character and properties induced by 
 the union of the elements to form these compounds respectively. 
 Hydrogen and chlorine, hydrogen and oxygen, hydrogen and 
 nitrogen, may be mixed in any proportion whatever, and their 
 respective properties still remain traceable in the mixtures ; but 
 in the compounds of hydrogen with chlorine, of hydrogen with 
 oxygen, of hydrogen with nitrogen, designated respectively hydro- 
 chloric acid, water, and ammonia, the elements are combined only 
 in the determinate proportions, and undergo only the determinate 
 changes of property, which we have experimentally verified. 
 
 This constancy of composition and character in chemical 
 compounds is a law to which there is no exception. 
 
 The immutability of the ratios in which the elements unite 
 to produce compound bodies is strikingly illustrated by the 
 formation of hydrochloric acid and water. 
 
 To demonstrate this fact, we will employ a glass tube, 
 divided by a glass stop-cock into two compartments of unequal 
 capacity, and having its open ends provided with glass stoppers 
 (Fig. 58). With this apparatus we will make two experiments. 
 
 In the first experiment we will fill the smaller compartment, 
 which is only about half the size of the larger, with dry hydro- 
 gen, the larger, with dry chlorine ; and after bringing the two 
 into communication by turning the stop-cock, we will expose 
 the apparatus to diffuse daylight. The two gases gradually 
 commingle, and soon begin to combine. The reaction is com- 
 pleted by exposing the tube for a short time to the influence of 
 the direct solar beams. On opening the tube under water, the 
 liquid is found to rise and fill a space double that of the smaller, 
 or hydrogen compartment. The residuary gas is chlorine ; the 
 volume of which, owing to its solubility in water, rapidly 
 diminishes. 
 
 la the second experiment we will fill the smaller compart- 
 ment with chlorine, the larger one with hydrogen, and allow 
 
EXPERIMENTAL DEMONSTRATION THEREOF. 
 
 61 
 
 the two gases to act upon one another under the influence of 
 light, as before. The volume of water which now rises in the 
 apparatus is equal to that previously observed ; but the re- 
 siduary gas is now hydrogen. 
 
 In these contrasted experiments, hydrogen and chlorine, 
 though mixed in reversed proportions, have nevertheless com- 
 bined in volumes absolutely unaltered ; the excess, whether of 
 the one or the other gas, taking no part in the reaction. 
 
 Fro. 53. 
 
 Analogous phenomena are witnessed during the formation of 
 water by the combination of hydrogen with oxygen. 
 
 To prove this experimentally, three U tubes, provided with 
 spark-wires and outlet-pipes, similar to those previously em- 
 ployed (comp. p. 51), are filled with mercury, and fixed in an 
 appropriate stand (Fig. 59), By the electrolysis of water in the 
 
62 CHANGES OF PROPERTY 
 
 apparatus already described (comp. p. 24), we obtain a mixture 
 of hydrogen and oxygen gases in the ratio in which they exist 
 in water. A column of this gaseous mixture, eight or nine 
 centimetres high, is allowed to rise in each of the three tubes ; 
 the height being marked by caoutchouc clipping-rings, after the 
 mercury has been rendered level in the two limbs of each appa- 
 ratus by the usual means. Into one of the tubes we pass, 
 moreover, an additional volume of hydrogen, into the other 
 an additional volume of oxygen ; these additional volumes 
 being likewise marked by means of caoutchouc rings. 
 
 Of the three U tubes, one in our experiment the middle 
 one now contains a mixture of hydrogen and oxygen in the 
 proportion in which the two gases are separated in the electro- 
 lysis of water ; the other contains the same mixture -f- an 
 additional volume of hydrogen; the third contains the same 
 
 FIG. 59. 
 
 \ 
 
 mixture -f- an additional volume of oxygen. The spark of the 
 induction-coil is now passed through the three mixtures, giving 
 rise in each of them to the formation of water. In the middle 
 tube every trace of gas is now found to have disappeared, the 
 mercury filling the tube to the very top ; in the second and 
 third tubes, volumes of gases remain, which, after appropriate 
 adjustment of the mercury in the two limbs, are found to be 
 equal respectively to the volume added in each case ; and these 
 gases, when liberated through stop-cocks adapted to the tubes 
 for that purpose, are respectively recognized, by means of the 
 usual tests, as hydrogen and oxygen. 
 
 Hence it appears that, in these three experiments, hydro- 
 
ACCOMPANYING CHEMICAL COMBINATION. 
 
 63 
 
 gen and oxygen did exclusively combine in the proportion in 
 which the two gases had been originally evolved from water. 
 The hydrogen and oxygen afterwards respectively admitted in 
 excess took no part in the formation of water. 
 
 The essential changes which the properties of the elements 
 undergo in associating to form chemical compounds are suffi- 
 ciently obvious from experiments with which we are already 
 familiar. 
 
 It would be impossible to quote a more instructive example 
 
 FIG. 59. 
 
 of such changes than that furnished by the transformation of 
 a mixture of hydrogen and chlorine into hydrochloric acid. 
 In the mixture of the two gases obtained by the electrolysis of 
 hydrochloric acid, the properties oi^each gas are easily re- 
 cognized. The mixture retains the fundamental property of 
 hydrogen ; it is still inflammable. The mixture also possesses 
 the yellowish-green colour of chlorine, its odour, and its bleach- 
 ing powers. On treating the mixture with water, its soluble 
 constituent is absorbed, its colour, odour, and bleaching powers 
 become weaker and weaker, until at last there only remains the 
 colourless, inodorous, tasteless hydrogen. Now let the mixture, 
 by one of the processes with which we have become acquainted, 
 
64 MECHANICAL MIXTURE CHEMICAL COMBINATION. 
 
 be converted into a chemical compound, and instead of the 
 yellowish-green, colour-bleaching, suffocating chlorine, sparingly 
 soluble in water instead of the inodorous, tasteless, inflammable 
 hydrogen we have a colourless gas, possessing no longer the 
 slightest bleaching power, absorbed by water with avidity, 
 of pungent odour and taste, and utterly incapable of com- 
 bustion. 
 
 In like manner, even a cursory comparison of the properties of 
 hydrogen and oxygen gases with those of the liquid water which 
 they produce by their union exhibits differences as striking as 
 can well be conceived ; but even in water-gas (dry steam), the 
 fundamental property of hydrogen (inflammability), and that 
 of oxygen (power of supporting combustion), are found to be 
 entirely extinct. 
 
 Not less striking is the change in the properties of hydrogen 
 and nitrogen, when associated to form ammonia. Two elemen- 
 tary gases, inodorous, insoluble in water, without action on 
 vegetal colours, are converted by chemical combination into a 
 gaseous compound, possessing a most penetrating odour, capable 
 of restoring acid-reddened litmus-paper to its pristine blue 
 colour, and so intensely and rapidly soluble in water, that 
 the fluid rushes into a tube filled with ammonia, as into a 
 vacuum. 
 
 The preceding remarks sufficiently mark the difference between 
 mechanical mixture and chemical combination. In mechanical 
 mixtures the elements are capable of interfusion in any propor- 
 tion whatever ; in chemical compounds they unite in definite 
 immutable proportions, volumetric and ponderal. The me- 
 chanical mixture exhibits properties intermediate between those 
 of its ingredients ; in the chemical compound the properties of 
 its constituents become extinct, their individuality being, so to 
 speak, merged in the formation of a new body, with new pro- 
 perties. 
 
 The recognition of these marked differences between mere 
 mechanical mixtures and definite chemical compounds very 
 naturally leads us to examine the conditions under which the 
 
CHEMICAL ACTION OF LIGHT AND HEAT. 65 
 
 former become converted into the latter. In this case, too, the 
 formation of hydrochloric acid and of water affords us welcome 
 elucidations. 
 
 A mechanical mixture of hydrogen and chlorine, as furnished 
 by the electrolysis of hydrochloric acid, may, if due care be 
 taken to protect it from the action of light, be preserved for an 
 indefinite period without undergoing the slightest change. 
 Under the influence of ordinary diffuse daylight, the trans- 
 formation of the mixture into a compound is accomplished, as 
 we have seen, in the course of a few hours. Direct sunlight, or 
 certain artificial lights of great intensity, effect the transition 
 instantaneously (comp. p. 46). Combination then takes place 
 with violent explosion, sometimes shattering the vessel which 
 contained the mixture. Transformation of the mixture into a 
 compound may in this case also be accomplished by contact with 
 a burning body, or by the passage of an electric spark ; both 
 which processes are attended by explosive violence. 
 
 The transformation of a mechanical mixture of hydrogen and 
 oxygen into the chemical compound, water, is less easily ac- 
 complished. It appears from recent experiments, that they 
 also may be caused to combine by sunlight though only on very 
 protracted exposure to its influence. On the approach of a 
 flame, however, or the passage of the electric spark, the mixture 
 instantaneously explodes, and is, at the same moment, converted 
 into water. 
 
 Hence it appears that mechanical mixtures are often trans- 
 formed into chemical compounds by the action of light, and, 
 more frequently still, by the powerful influence of heat. The 
 experience acquired in the study of hydrochloric acid and 
 water cannot, unfortunately, be extended by the similar exami- 
 nation of ammonia, seeing that no process is known for directly 
 effecting the transformation of a mixture of hydrogen and 
 nitrogen into a compound thereof. But this exception renders 
 it the more necessary to mention that the conditions which 
 determine the formation of hydrochloric acid and water from 
 mixtures of their elements suffice for the accomplishment of 
 
.* 
 66 CHEMICAL EVOLUTION OF LIGHT AND HEAT. 
 
 like results in an endless variety of cases ; heat being especially 
 distinguished as the most general promoter of chemical trans- 
 formation. 
 
 When proceeding hereafter with the special study of the 
 individual elements, we shall have frequently to resume the 
 consideration of these conditions of chemical change. Oppor- 
 tunities will then also occur for examining, minutely, certain 
 interesting concomitants of chemical activity; such as, for 
 instance, the remarkable development of light and heat which 
 we have already seen attending the reactions in several of our 
 illustrative experiments. 
 

 
 LECTUliE IV. 
 
 1 
 
 Chemical symbols their *fn'tttiu uad f tittle* -diagrammatic symbols 
 initialed figured symbolic equations constructed therewith informa- 
 tion thereby conveyed formulae thence derived summary of information 
 condensed in chemical formulae hydrochloric acid, water, and am- 
 monia, considered as types of chemical combination hydrobromic and 
 hydriodic acids their construction upon the type of hydrochloric acid 
 ponderal analysis and volume-weights of these compounds volume- 
 weights of bromine and iodine gases sulphuretted and selenetted hydro- 
 gen their construction upon the type of water ponderal analysis and 
 volume-weights of these compounds volume-weights of sulphur and 
 selenium gases. 
 
 THE principal facts determined by experiment, as to the com- 
 position of hydrochloric acid, water, and ammonia, are sus- 
 ceptible of clear and concise expression in a few happily- 
 chosen symbols, which experience has shown to be so powerful 
 as instruments of chemical research, and so invaluable as 
 adjuncts of chemical nomenclature, that they may justly claim 
 precedence over all the subjects pressing for attention at this 
 early stage of our inquiry. 
 
 If equal volumes of hydrogen, chlorine, oxygen and nitrogen 
 (taken, of course, at like temperature and pressure) be repre- 
 sented by equal squares, having the initials of these elements 
 inscribed therein, the composition, by volume, of hydrochloric 
 acid, water, and ammonia, may be thus expressed : 
 
 Hydrochloric Acid. 
 
 Water. 
 
 Ammonia. 
 
 H 
 
 01 
 
 H 
 
 
 H 
 
 N 
 
 H 
 
 
 By inscribing within these symbolic squares the volume-weights 
 
68 
 
 CHEMICAL SYMBOLS FIGURED, LETTERED, 
 
 of the elements instead of the initial letters of their names, we 
 obtain a series of figured symbols, which, when placed side by 
 side with the lettered ones, furnish the following instructive 
 equations, exhibiting on one hand the volumes, and on the other 
 hand the weights, of the elements entering into combination : 
 
 Hydrochloric Acid. 
 
 Water. 
 
 H 
 
 Cl 
 
 1 
 
 35.5 
 
 H 
 
 
 
 
 H 
 
 
 Ammonia. 
 
 H 
 
 
 H 
 
 N 
 
 II 
 
 
 1 
 
 
 1 
 
 14 
 
 1 
 
 
 Again, if we add together the weights in the figured symbols 
 respectively, thus : 
 
 1 + 35-5 - 36-5 
 
 1 + 1 + 16 = 18 
 
 1 + 1 + 1 + 14 =17 
 
 we learn the composition, by weight, of 36'5 parts of hydro- 
 chloric acid, of 18 parts of water, and of 17 parts of ammonia 
 respectively. From these ratios, the centesimal composition of 
 each compound is, of course, deducible by the well-known rule 
 of proportion. Thus with respect to hydrochloric acid : 
 
 36-5 : 1-100 : x 
 
 100 
 x = = 2-74 per cent, of hydrogen, 
 
 OO'O 
 
AND VOLUMETRIC. 69 
 
 and 100 - 2-74 = 97'26 per cent, of chlorine, 
 forming 100-00 parts of hydrochloric acid. 
 
 So, again, with respect to water : 
 
 18 : 2 = 100 : y 
 
 y= . = 11*11 per cent, of hydrogen, 
 
 18 
 
 and 100 - 11-11 = 88-89 per cent, of oxygen, 
 forming 100- 00 parts of water. 
 
 And so, lastly, with respect to ammonia : 
 
 17 : 3-100 : z 
 
 300 
 z = - - 17-64 per cent, of hydrogen, 
 
 and 100 17-64 = 82-36 per cent, of nitrogen, 
 forming 100 -00 parts of ammonia. 
 
 By connecting, in our memory, the volume-weights of the 
 elements with their respective initials, we have all the preceding 
 information brought before us in the three simple symbolic 
 formulae given at the outset of this Lecture. 
 
 We must, however, proceed a step further, in order to obtain 
 an equally clear and concise symbolic representation of the 
 relative volume-weights (or specific gravities) of the three 
 compounds under review hydrogen being taken as unity. 
 
 For this purpose, we must bear in mind the experiments by 
 which we established, firstly, that hydrogen and chlorine com- 
 bine to form hydrochloric acid without condensation ; secondly, 
 that, during the union of hydrogen and oxygen, the volume 
 of the water-gas formed is only two-thirds of the volume of 
 the mixed gases ; and thirdly, that the formation of ammonia 
 involves the condensation of the constituents to one-half 
 of their volume before combination. These facts may be gra- 
 phically symbolized as follows : 
 
70 
 
 UNIT-VOLUMES PRODUCT-VOLUMES. 
 
 Hydrochloric Acid. 
 
 1 vol. 4- 1 vol. = 
 Water. 
 
 2 vols. 
 
 2 vols. 4- 1 vol. = 
 Ammonia. 
 
 2 vols. 
 
 H 
 
 1 
 
 
 
 
 H 
 1 
 
 4 14 HaN 17 
 
 
 
 H 
 1 
 
 
 3 vols. 
 
 + 1 vol. = 
 
 2 vols. 
 
 These symbolic equations show, at a glance, the fact (of 
 which we have already had experimental proof) that very un- 
 equal weights of these compounds occupy, under the same 
 conditions of temperature and pressure, exactly equal spaces ; 
 the space occupied, in each case, being exactly double the space 
 filled by the unit-volume of hydrogen. In other words, the 
 volume-weight of hydrogen being 1 
 
 The double volume-weight of hydrochloric acid gas is 36*5 
 of water gas, or dry steam 18 
 
 3 , of ammonia gas . . 17 
 
 Evidently, therefore, we have only to halve these double volume- 
 weights of hydrochloric acid, water-gas, and ammonia, in order 
 to arrive at their respective single volume-weights, relatively 
 
INFORMATION EMBODIED IN SYMBOLS. 71 
 
 to hydrogen taken as unity ; in other words, at their respective 
 specific gravities : thus, 1 being the volume-weight of hydrogen, 
 we have : 
 
 Ofl.K. 
 
 - = 18-25 = the relative vol.- weight or sp.gr. of hydrochloric acid. 
 
 2i 
 
 18 
 
 = 9 of water-gas. 
 
 17 
 
 = 8*5 = of ammonia. 
 
 When these ratios of volume, weight, and condensation, have 
 been once firmly fixed in the memory by aid of the figured 
 squares, this graphic or diagrammatic form of symbolism may 
 be dispensed with, and the whole of our experimentally acquired 
 knowledge of hydrochloric acid, water, and ammonia, may be 
 embodied in the following succinct expressions : 
 
 HC1 = 2 vols. 
 
 jjo or H 2 = 2 vols. 
 
 H| 
 
 HVN or H 3 N = 2 vols. 
 Hj 
 
 The importance of these concise formulae is best evinced by a 
 simple recapitulation of the information they afford at a glance. 
 We learn from them : 
 
 1. The number and names of the elements entering into the 
 composition of hydrochloric acid, water, and ammonia respec- 
 tively. 
 
 2. The ratios in which the elements are united in these 
 several compounds by volume. 
 
 3. The ratios in which the elements are united therein by 
 weight. 
 
 4. The ratios in which the volumes of the several compounds, 
 when formed, stand to the volumes of their respective con- 
 stituents before combination. 
 
72 CONGENERS OF CHLORINE. 
 
 5. The relative volume-weights or specific gravities of these 
 several compounds in the state of gas or vapour, hydrogen being 
 taken as unity. 
 
 The weight and volume ratios represented by the formulae of 
 hydrochloric acid, water, and ammonia, 
 
 HC1, H 2 0, and H 3 N, 
 
 are not merely interesting in themselves, but acquire a still 
 deeper significance when it is found that they are types or 
 models, representing as many groups of compounds, each cast, 
 so to speak, in the same mould, and governed by the same 
 structural law, as its generic prototype. These analogous bodies 
 come next in order for consideration. 
 
 In our previous studies, devoted to the prototypes, we have 
 adduced experimental proof of each one of our propositions. 
 Similar evidence exists as to the structure of the analogues, 
 which we have now to review. It would, however, encumber 
 our present demonstration, and lead us too far from our im- 
 mediate purpose, were we to attempt, at this stage of our 
 progress, the experimental verification of every fact adduced 
 in support or extension of our arguments. In the pursuit of 
 scientific knowledge, much must at all times be admitted on 
 adequate testimony ; and we, like other students of nature, must 
 refer for many of our facts to the recognized stores of science, 
 reserving for a future period the experimental demonstration of 
 some of the results in the meanwhile taken on trust. 
 
 Among the elementary bodies, which we shall hereafter have 
 to investigate, there are two, named respectively Bromine and 
 Iodine, whose properties, as we shall find, ally them closely 
 with chlorine. Like chlorine, for example, these two bodies 
 unite with hydrogen, producing gaseous compounds respectively 
 called hydrobromic and hydriodic acids, each strikingly analo- 
 gous to hydrochloric acid, and manifestly constructed upon its 
 type. 
 
 Again, there are two elements called respectively Sulphur and 
 
CONGENERS OF OXYGEN AND NITROGEN. 73 
 
 Selenium, presenting many points of analogy with oxygen ; and, 
 amongst other things, forming, with hydrogen, gaseous com- 
 pounds called respectively sulphuretted and selenetted hydro- 
 gen, which, though differing in many of their properties from 
 the compound of hydrogen with oxygen, are nevertheless 
 chemical congeners with water, as being unquestionably con- 
 structed upon the same type. 
 
 So, once more, do the elements Phosphorus and Arsenic 
 present themselves as analogues of nitrogen, when studied in 
 their combinations with hydrogen, forming the gaseous com- 
 pounds termed respectively phosphoretted and arsenetted hydro- 
 gen. At the first glance these hydrogen-compounds present 
 but few points of resemblance with ammonia ; but closer in- 
 vestigation reveals a construction so directly comparable, that it 
 is impossible not to recognize, in phosphoretted and arsenetted 
 hydrogen, the reproduced type of ammonia. 
 
 Let us endeavour to trace, step by step, the considerations 
 which have led chemists to the admission of these typical rela- 
 tions. 
 
 If the view of the analogous construction of hydrochloric, 
 hydrobromic, and hydriodic acids be correct, 2 vols. of hydro- 
 bromic and hydriodic acid gases must contain 1 vol. of hydrogen 
 each, united with 1 vol. of bromine-gas and 1 vol. of iodine-gas 
 respectively. Now, this is exactly the result which has been 
 established by experiment; though the method employed to 
 verify the composition of hydrobromic and hydriodic acids 
 differs from that which we found it convenient to adopt in the 
 case of hydrochloric acid. 
 
 The circumstances necessitating this change of method may 
 be easily explained. At ordinary atmospheric temperatures 
 chlorine is a gas; and this body consequently lends itself 
 readily to the volumetric determinations which we adopted in 
 its study. But it is far otherwise with bromine and iodine. At 
 ordinary atmospheric temperature the former body is a liquid, 
 the latter a solid body. Hence it follows that the volumetric 
 
74 PONDEKAL DETERMINATION OF THE VOLUME-RATIOS 
 
 determination of the quantity of bromine-gas and iodine-gas 
 contained in hydrobromic and hydriodic acids respectively 
 would be attended with considerable difficulty. For this reason 
 it is found preferable, in studying the composition of these 
 substances, to replace the volumetric by the ponderal method 
 of measurement. Indeed, the analogy of hydrobromic and hy- 
 driodic acids with hydrochloric acid may be quite as satisfac- 
 torily determined by means of the balance, as by direct volumetric 
 analysis. 
 
 It is here that we avail ourselves for the first time of 
 this indirect method of ascertaining volumetric relations. As 
 we proceed, however, we shall see that in most instances the 
 balance is our safest and readiest guide to accurate volumetric 
 determinations ; while in many cases it is the only means avail- 
 able for use in such inquiries. Hence the method of ascertain- 
 ing volume-ratios by ponderal analysis deserves, even now, our 
 attentive consideration. It would, however, lead us too far from 
 our immediate subject were we here to enter upon an inquiry 
 into the nature of ponderal analysis, and the means, sometimes 
 simple, oftener very complex and refined, by which the chemist 
 eliminates and weighs in succession the several constituents of a 
 compound body. Nor can we make this method of investiga- 
 tion the subject of rapid experimental demonstrations, such as 
 are alone suited for our purpose during the brief periods of our 
 assembling here to pursue our inquiries together, and such as we 
 found ourselves able to employ 'in our study of the volumetric 
 method. Ponderal analysis proceeds for the most part by 
 operations too numerous, too protracted, and too nice for per- 
 formance, in the short time, and with the restricted appliances, at 
 the lecturer's disposal. We must, therefore, reserve for future 
 occasions the experimental study of the methods employed in 
 ponderal analysis ; contenting ourselves for the present with its 
 results, and selecting from these such as bear directly upon the 
 subject we have in hand. 
 
 Our present inquiry relates to the three compounds, hydro- 
 chloric, hydrobromic, and hydriodic acids; and the ponderal 
 
OF HYDKOCHLOEIC, HYDROBROMIC, AND HYDRIODIC ACIDS. 75 
 
 analysis of these three compounds has elicited the following 
 results : 
 
 In hydrochloric acid'j f 3 5'5 parts of chlorine. 
 
 TIT! -! 1 part of hydrogen OA , 
 
 In hydrobromic acid ;/ combin ^ w f th bromme. 
 
 In hydriodic acid J 1127 iodine. 
 
 In what manner can these ponderal results be made available 
 to display the volumetric construction of hydrobromic and 
 hydriodic acids in comparison with that of hydrochloric acid ? 
 
 We have seen that 
 
 1 part of hydrogen + 35'5 parts of chlorine = 36*5 parts 
 of hydrochloric acid. 
 
 And we have also seen that the 36 '5 parts of hydrochloric 
 acid thus obtained occupy exactly double the space filled by the 
 unit-volume of hydrogen. The value 36 '5 accordingly repre- 
 sented for us the double volume-weight of hydrochloric acid ; of 
 which, consequently, the single volume- weight, i.e., the volume- 
 
 Q K 
 
 weight relatively to hydrogen, or the sp. gr., is . = 18'25. 
 
 Supposing hydrobromic and hydriodic acids to be similarly 
 constructed, we must have 
 
 1 part of hydrogen + 80 parts of bromine = 81 parts of 
 hydrobromic acid = 2 vols. ; whence the volume-weight of 
 
 81 
 hydrobromic acid -- = 4O5 ; and 
 
 1 part of hydrogen + 127 parts of iodine = 128 parts of 
 hydriodic acid = 2 vols. ; whence the volume- weight of 
 
 hydriodic acid = = 64. 
 
 2 
 
 The experimental determination of the respective volume- 
 weights has justified these previsions. 
 
 We remember, lastly, that the weight of chlorine (35*5) 
 which combines with 1 volume-weight of hydrogen, to form 2 
 volumes of hydrochloric acid, represents the volume-weight of 
 
 E 2 
 
76 
 
 DIAGEAMMATIC EQUATIONS OF STRUCTURE. 
 
 chlorine. Assuming the hydrobromic and hydriodic acids to 
 have an analogous constitution, it is obvious that the weight of 
 bromine (80) and the weight of iodine (127) respectively united 
 with 1 volume-weight of hydrogen to form 2 volumes of each 
 acid, must represent the respective volume-weights of bromine and 
 iodine in the state of gas ; or, briefly, their gas-volume-weights, 
 often termed their vapour-densities. The comparison of these 
 elements as gases must of course be conducted at temperatures 
 equal or superior to that at which the least volatile of them 
 assumes the gaseous form ; and experiment has proved that, at 
 equal temperatures, bromine gas is 80 times, and iodine gas 127 
 times, heavier than hydrogen gas. 
 
 These comparisons can leave no doubt that hydrobromic and 
 hydriodic acids are respectively constructed upon the type of 
 hydrochloric acid ; and, if we represent the respective volume- 
 weights of bromine gas and iodine gas byBr = 80 and I = 127, 
 our information regarding the three analogously-constructed 
 compounds may be thus symbolically epitomized : 
 
 Composition of hydrochloric acid. 
 
 II 
 
 Composition of hydrobromic acid. 
 
 Composition of hydriodic acid. 
 
 Or, replacing the diagrammatic symbols, as before, by succinct 
 formulae, we have 
 
 For hydrochloric acid H + Cl. = HC1. 
 For hydrobromic acid H -f Br = HBr. 
 For hydriodic acid H + I = HI. 
 
WATER-GAS, SULPHURETTED AND SELENETTED HYDROGEN. 77 
 
 On comparing the composition of sulphuretted and selenetted 
 hydrogen with that of water, we arrive at analogous results, 
 capable of similar diagrammatic expression. 
 
 Two volumes of water-gas, as we have seen, contain 2 vols. of 
 hydrogen united with 1 vol. of oxygen. If sulphuretted and 
 selenetted hydrogen be constructed on the same type, we must 
 find, in 2 vols. of each of these compounds, 2 vols. of hydrogen 
 united with 1 vol. of sulphur-gas and 1 vol. of selenium-gas 
 respectively. Experiment has verified this anticipation. But, 
 as both sulphur and selenium are solids at the common tempera- 
 ture, becoming gases only under the influence of powerful heat, 
 chemists have preferred ponderal to volumetric analysis for the 
 investigation of these cases also ; and the balance has elicited 
 the following results : 
 
 In water \ 2 parts of f 16 parts of oxygen. 
 
 In sulphuretted hydrogen > hydrogen are < 32 sulphur. 
 In selenetted hydrogen J combined with (79.4 j? selenium. 
 
 From these results the gas-volume-weights of the three com- 
 pounds are readily computed as follows : 
 
 For water-gas 
 
 2 parts of hydrogen + 16 parts of oxygen = 18 parts of 
 water = 2 vols. ; whence the volume-weight of water-gas = 
 
 For sulphuretted hydrogen 
 
 2 parts of hydrogen + 32 parts of sulphur = 34 parts of sul- 
 phuretted hydrogen = 2 vols. ; whence the volume- weight of 
 
 sulphuretted hydrogen = - =17. 
 
 2i 
 
 For selenetted hydrogen 
 
 2 parts of hydrogen -f 79 parts of selenium = 81 parts of 
 selenetted hydrogen = 2 vols. ; whence the volume- weight of 
 
 81 
 
 selenetted hydrogen = = 40*5. 
 2 
 
78 
 
 ANALOGOUS COMPOSITION OP 
 
 The two last results, calculated on the assumption that sul- 
 phuretted and selenetted hydrogen are constructed on the type of 
 water, coincide exactly with the results obtained by experiment. 
 
 But our previous inquiries (comp. p. 50) have also taught us 
 that the weight of oxygen (16), which is combined, in water-gas, 
 with 2 parts (i.e., with the weight of 2 vols.) of hydrogen, repre- 
 sents the volume-weight of oxygen. If sulphuretted hydrogen 
 and selenetted hydrogen be analogously constituted, we are 
 justified in expecting that the weights of sulphur (32), and 
 of selenium (79), which enter into combination with 2 parts 
 of hydrogen, express also the respective volume-weights of 
 the sulphur and selenium gases. Quite recently the volume- 
 weights of sulphur and selenium gases have been accurately de- 
 termined by experiment ; and the results show that, at tem- 
 peratures at which these elements assume the perfectly gaseous 
 condition, sulphur-gas is 32 times, and selenium-gas 79 times, 
 heavier than hydrogen. 
 
 The analogy of the construction of water, of sulphuretted, and 
 of selenetted hydrogen is thus most satisfactorily made out ; 
 and, if the volume-weights of sulphur-gas and selenium-gas be 
 respectively represented by S = 32, and Se = 79, the volumetric 
 information afforded, in these cases, by ponderal analysis, may 
 be thus diagrammatically symbolized : 
 
 Composition of water-gas. 
 
 Composition of sulphuretted hydrogen gas 
 
WATER-GAS, SULPHURETTED AND SELENETTED HYDROGEN. 79 
 
 Composition of selenetted hydrogen gas. 
 
 Se 
 
 Dropping, as before, our diagrammatic squares, we obtain tho 
 following concise expressions : 
 
 For water-gas 2H + = H 2 O. 
 
 For sulphuretted hydrogen 2H + S = H 2 S. 
 For selenetted hydrogen 2H + Se = H 2 Se. 
 
 It now only remains for us to ascertain, in the last place, 
 whether the typical construction of ammonia be really repro- 
 duced, in conformity with our anticipation, in phosphoretted 
 and arsenetted hydrogen ; but this inquiry, together with some 
 general remarks on chemical symbolization, must be reserved 
 for a separate lecture. 
 
80 
 
 LECTUKE V. 
 
 Chemical symbols (continued) phosphoretted and arsenetted hydrogen 
 their construction upon the type of ammonia ponderal analysis and 
 volume- weights of these compounds exceptional volume-weights of 
 phosphorus and arsenic gases combining weights of phosphorus and 
 arsenic general remarks on chemical symbolization chemical formulae 
 as instruments of classification representation of chemical processes in 
 equations translation of formulae into concrete weights and volumes 
 ponderal analysis of sodic and potassic chlorides, oxides, and nitrides 
 determination of the combining weights of sodium and potassium. 
 
 HAVING considered the structure of the compounds included in 
 two of our typical groups, those, namely, which are formed in 
 the mould of hydrochloric acid and water, we have now lastly 
 to examine, in like manner, our third typical group, that, to wit, 
 at the head of which we have placed ammonia ; our object being 
 to ascertain whether the structural type of this body is really 
 reproduced, as we conceive it to be, in phosphoretted and 
 arsenetted hydrogen. 
 
 For this purpose we must bear in mind that, while 2 vols. of 
 hydrochloric acid were found to contain 1 vol. of hydrogen com- 
 bined with 1 vol. of chlorine ; and while 2 vols. of water-gas were 
 found to contain 2 vols. of hydrogen combined with 1 vol. of 
 oxygen ; we established that 2 vols. of ammonia contain 3 vols. of 
 hydrogen united with 1 vol. of nitrogen. The question now 
 before us is, whether phosphoretted and arsenetted hydrogen 
 tally precisely with this last-mentioned type ? 
 
 Here, again, chemists have had recourse to ponderal analysis 
 for the desired information ; and the composition of ammonia, 
 phosphoretted and arsenetted hydrogen, as respectively deter- 
 mined by aid of the balance, is as follows : 
 
 In ammonia 
 
 In phosphoretted hydrogen 
 
 In arsenetted hydrogen 
 
 3 parts of 
 hydrogen 
 are com- 
 bined with 
 
 14 parts of nitrogen. 
 31 phosphorus. 
 75 arsenic. 
 
AMMONIA PHOSPHORETTED AND ARSENETTED HYDROGEN. 81 
 
 We know, from our previous experiments, that, in the case 
 of ammonia, these weight-results correspond to the relative 
 volume-weight, or specific gravity, subjoined : 
 
 3 parts of hydogren + 14 parts of nitrogen = 17 parts of 
 ammonia = 2 vols. ; whence the volume-weight or sp. gr. of 
 
 17 Q* 
 
 ammonia = = b'o, 
 u 
 
 If, therefore, the volumetric structure of phosphoretted and 
 arsenetted hydrogen be identical with that of ammonia, we ought 
 to obtain, as their relative volume-weights, the following respec- 
 tive values : 
 
 For phosphoretted hydrogen : 
 
 3 parts of hydrogen +31 parts of phosphorus = 34 parts of 
 phosphoretted hydrogen = 2 vols.; whence the volume- 
 weight of phosphoretted hydrogen = -- = 17. 
 
 2 
 
 For arsenetted hydrogen : 
 
 3 parts of hydrogen -{- 75 parts of arsenic = 78 parts of ar- 
 senetted hydrogen = 2 vols. ; whence the volume-weignt of 
 
 78 
 arsenetted hydrogen = = 39. 
 
 These are, in very truth, the volume-weights of phosphoretted 
 and arsenetted hydrogen, as furnished by experiment. 
 
 So far, therefore, as our inquiry bears upon the three compounds 
 in question, when already formed, and not upon their elements 
 while as yet uncombined, experiment seems to justify us in affirm- 
 ing the entire identity of the structural type exemplified in these 
 three cases. 
 
 But when, from the compounds formed, we turn to consider 
 the volumes of the elements which take part in their formation, 
 we discover a most remarkable and curious discrepancy the 
 first of its kind that our studies have brought under our notice. 
 To the nature of this discrepancy we must now pay particular 
 attention. 
 
 E 3 
 
02 EXCEPTIONAL VOLUME-WEIGHTS 
 
 The weight of nitrogen (14) which combines with 3 parts by 
 weight of hydrogen to form ammonia, is, as we have seen, the 
 volume-weight of nitrogen. In other words, the combining 
 weight of nitrogen coincides with its volume-weight. Assum- 
 ing the chemical construction of phosphoretted and arsenetted 
 hydrogen to be strictly analogous to that of ammonia, we should 
 expect that the weights of phosphorus (31) and of arsenic (75) 
 united with 3 parts of hydrogen, in phosphoretted and arse- 
 netted hydrogen, would represent the volume-weights of phos- 
 phorus and arsenic respectively. In other words, we should 
 expect, in their case, the same coincidence of the combining weight 
 with the volume-iveight as obtains in the case of nitrogen. 
 
 Here, however, we meet with the first exceptions to a rule 
 hitherto unbroken. 
 
 The volume-weights of phosphorus and arsenic are not re- 
 presented by the figures (31 and 75), which express their 
 respective combining weights ; though, as we shall presently 
 find, the two values stand in a very simple ratio to each 
 other. Upon experimentally comparing the weights of equal 
 volumes of hydrogen, phosphorus, and arsenic, at the tem- 
 perature at which these latter bodies (usually solid) become 
 gaseous, we find that phosphorus gas is not 31 times, nor arse- 
 nic gas 75 times heavier than hydrogen, but that each of these 
 figures has to be doubled to bring it into conformity with fact. 
 In other words, the volume-weight of phosphorus gas is not, 
 like its combining weight, 31, but 31 x 2 = 62 ; and the 
 volume-weight of arsenic gas is not, like its combining weight, 
 75, but 75 x 2 = 150. 
 
 This striking and singular deviation from a coincidence 
 hitherto constantly observed, stands before us at present unex- 
 plained. We are unable to suggest any end likely to be served 
 by these exceptional volumetric relations of phosphorus and 
 arsenic gases, and by the pro tanto deviation from the ammonia 
 type thus occasioned in phosphoretted and arsenetted hydrogen. 
 But though the purpose of this difference is as yet unknown to 
 us, its nature and limits are most clearly made out. 
 
OP PHOSPHORUS AND ARSENIC. 83 
 
 Each of the three compounds under consideration contains, in 
 two volumes, three volumes of hydrogen ; each also has, on ex- 
 periment, furnished a volume-weight, corresponding strictly 
 with that calculated from its ponderal analysis ; but the quanti- 
 ties of nitrogen, phosphorus, and arsenic combined, in the three 
 compounds, with three parts of hydrogen, are the weights of 
 unequal volumes of the three elements : the weight of the nitro- 
 gen (14) representing one volume of nitrogen, while the weights 
 of the phosphorus (31) and of the arsenic (75) represent only 
 half a volume of phosphorus and arsenic gases respectively. 
 These preponderant analogies, and this partial discrepancy, 
 observed in the construction of the three compounds, are con- 
 spicuously displayed in the following diagrams, which, it will be 
 observed, are of two kinds ; or, rather, which may be said to 
 depict the facts from two points of view. 
 
 In the first series the volume-weights of phosphorus and arsenic, 
 which, as we have just seen, are exactly double their combining 
 weights, are taken as the starting-point of the symbolic expres- 
 sion ; and the depicted volumes of phosphorus and arsenic are 
 thus brought into conformity with the unit-volume of hydrogen 
 our standard or normal volume. This involves, of course, the 
 duplication, as well of the three volumes of hydrogen taking 
 part in the combination, as of the normal product-volume (two 
 unit-volumes) of the compound in each case generated. It 
 also necessitates the adoption of two literal symbols in each 
 case, one to represent the volume-weight, the other to denote the 
 combining weight of the body in question. For phosphorus we 
 may conveniently employ, to denote these two weights respec- 
 tively, the symbols Pho = 62, and P = 31 ; for arsenic, the sym- 
 bols Ars = 150, and As = 75. The diagrammatic representations, 
 thus brought into conformity with fact, assume the following 
 appearance relatively to ammonia, which is prefixed, as a stand- 
 ard for comparison. Taking first the volume-weights as the 
 bases of the symbolic representations, we have : 
 
84 DIAGRAMS OF THE STRUCTURE OF AMMONIA, 
 
 Composition of ammonia. 
 
 J 
 
 Composition of phosphoretted hydrogen. 
 
 1 
 
 > 4- Pho 
 
 Composition of arsenetted hydrogen. 
 
 + Ars 
 
 Dropping the squares, as before, we obtain the following 
 formulae : 
 
 For ammonia 3H + N = H 3 N. 
 
 For phosphoretted hydrogen 6H + Pho. = H G Pho. 
 For arsenetted hydrogen 6H + Ars. = H 6 Ars. 
 
 In the second series of diagrams an opposite starting-point is 
 adopted. 
 
AND OF PHOSPHORETTED AND ARSENETTED HYDROGEN. 85 
 
 Composition of ammonia. 
 
 Composition of phosphoretted hydrogen. 
 
 Composition of arsenetted hydrogen. 
 
 It is not the volume-weights, but the combining weights of 
 phosphorus and arsenic that are here taken as the bases of the 
 respective symbolic representations. Consequently, the volumes 
 of phosphorus and arsenic gases are depicted as of only half the 
 size of the normal unit- volume. This form of diagram obvi- 
 ates the necessity of duplicating the hydrogen volumes and 
 the product-volumes, and gives a series of simpler symbols, in 
 which (as above mentioned) P stands as = 31, and As as = 75 ; 
 their semi-volumetric relation being represented by a special 
 symbol having only half the area of the square. 
 
86 VALUE OF CHEMICAL SYMBOLIZATION 
 
 Omitting the squares in this, as in the previous case, the formula 
 become : 
 
 For ammonia 3H + N = H 3 N. 
 
 For phosphoretted hydrogen 3H + P = H 3 P. 
 For arsenetted hydrogen 3H -f As = H 3 As. 
 
 It will, of course, be clearly understood that these two modes 
 of symbolization may be employed indifferently; and that 
 neither of them gives us any further insight into the chemical 
 structure of these two compounds, than that which is afforded 
 by the experimental results equally recorded in both forms of 
 diagram. The form exemplified in the first series may be de- 
 scribed as the bi-ponderal form ; while that displayed in the 
 second series may be designated the semi-volumetric. The latter 
 form is much to be preferred, as retaining unaltered, in each 
 case, the three unit-volumes of hydrogen, and the normal double- 
 volume of the resultant compound ; while the changed appear- 
 ance of the middle term in the formulae of phosphoretted and 
 arsenetted hydrogen fixes attention, at the first glance, on the 
 real and only point of difference between these formulae and 
 the normal ammonia-formula, viz., the semi-volumetric character 
 of phosphorus and arsenic gases, as compared with nitrogen. 
 
 We shall have to return on future occasions to the exceptional 
 characters of these phosphorus and arsenic compounds ; for the 
 present we commit to memory the experience that, while the 
 symbols H, Cl, Br, I, 0, S, Se, and N, respectively express both 
 the combining weights and the volume-weights of hydrogen, chlorine, 
 bromine-gas, iodine-gas, oxygen, sulphur-gas, selenium-gas, and 
 nitrogen, the symbols P and As express the combining, but not the 
 volume-weights of phosphorus and arsenic gases ; the latter being 
 double the former weight in the case of each of these elements. 
 
 Though the complete unity of the symbolic notation is dis- 
 turbed by these two exceptional cases, the practical value and 
 importance of this admirable language which may be well 
 termed the Algebra of chemistry remains entirely unimpaired. 
 
AS AN INSTRUMENT OF RESEARCH AND RECORD. 87 
 
 By replacing, for phosphorus and arsenic, the square which ex- 
 presses a full volume, by the triangle resulting from the diagonal 
 bisection of that square, the semi-volumetric character of these 
 exceptional bodies is appropriately symbolized (as in the above 
 diagram), and all the other symbolic expressions fall under a 
 universal rule. 
 
 Of the value of such symbolic formulae we have already had 
 many examples ; and if they possessed no other capability than 
 that, already known to us, of tersely embodying the elementary 
 structure of particular compounds, they would deserve to rank 
 among the chemist's most powerful instruments of research. 
 
 Instructive, however, as individual formulae are shortly as 
 they sketch, so to speak, the portrait of each compound, and 
 vividly as they impress its salient features on the mind their 
 value becomes still more conspicuous when they are studied in 
 contrast with each other, and employed as instruments of classifi- 
 cation. By comparing the formula of any given compound with 
 each of the typical formulae in succession, the group to which 
 the compound belongs is speedily revealed ; and from the known 
 chemical characters of the group, the probable properties of a 
 newly -introduced member may be forecast, so as to shape out an 
 appropriate course for research, and not seldom to anticipate 
 its results. 
 
 But it is, perhaps, after all, as a language as an instrument of 
 record, rapidly written, rapidly read, and capable of presenting 
 collectively to the mind extensive concatenations of fact that 
 words could only convey piecemeal it is probably in this, which 
 may be termed its short-hand application, that the value of 
 chemical formulas is, in the practical cultivation of the science, 
 most conspicuously manifested. 
 
 Nor is it for practical purposes only that chemical formulae 
 are thus valuable. In the hands of the chemical philosopher 
 they become the expressions, at once concise and comprehen- 
 sive, of chemical laws the abstract representations of definite 
 proportionalities which are transformed into statements of 
 actual facts, when the symbols are replaced by the concrete 
 values for which they stand. 
 
88 CHEMICAL EQUATIONS EXEMPLIFIED I 
 
 For all these purposes chemical symbols may be handled like 
 ordinary algebraic expressions. The symbolic representations 
 of the various elements and compounds may be connected by 
 the ordinary algebraic signs : by the sign of addition, + ; of 
 multiplication, x ; of subtraction, ; and of equality, = . 
 
 In this manner we may construct Chemical Equations, by aid 
 of which the most complicated chemical reactions may be 
 rapidly traced out through all their conceivable permutations ; 
 each possible chemical change thus set forth suggesting, per- 
 haps, an experiment. On the other hand, the worthlessness of 
 particular schemes of research may be often perceived before- 
 hand, or the fallacies of sonxe too hasty generalization detected 
 without difficulty, by merely subjecting the anticipated reactions 
 to the searching test of a symbolic investigation. 
 
 The young chemist cannot, indeed, too early, nor too earnestly, 
 study the language of chemical formula. Long before he is 
 far enough advanced in his studies to appreciate their higher 
 philosophical uses, he will find them invaluable aids of his pro- 
 gress, if only to test the clearness and precision of his reason- 
 ings, and the accuracy of his experimental results. Results 
 may, indeed, happen to be susceptible of expression in logical 
 symbolic equations, without, on that account, deserving confi- 
 dence as of necessity accurate ; but those which cannot be 
 brought into coincidence with the appropriate formulae may, with 
 perfect safety, be set aside as inexact. It should, therefore, be 
 the student's object, from the outset of his career, to attain 
 fluency both in reading and writing chemical symbols and 
 formulae ; so that he may be prepared, at a later period, to acquire 
 skill in the art of applying them, for the solution of practical 
 and theoretical problems. 
 
 As at once an introduction to this study, and a summary of 
 the reactions which we have already passed in review, we will 
 proceed to record these, more fully and precisely than before, 
 by availing ourselves of the language of chemical formulaB. 
 
 The student doubtless bears in mind how useful we found 
 the strong attraction of chlorine for hydrogen, in revealing to 
 
DECOMPOSITION OF WATER BY CHLORINE. 89 
 
 us the composition of water and ammonia ; and how readily, 
 by means of this powerful agency, we were enabled to with- 
 draw hydrogen from both those compounds, so as to liberate 
 therefrom the oxygen and nitrogen they respectively contain. 
 The formation of hydrochloric acid on the one hand, and the 
 liberation of oxygen and nitrogen respectively on the other, 
 place beyond doubt the qualitative composition of the two com- 
 pounds thus examined. 
 
 We must go further than this, however, in order to arrive at a 
 symbolic statement of these changes. We must obtain a quan- 
 titative as well as a qualitative knowledge of the reactions which 
 take place in these experiments. These investigations have been 
 made. Chemists have determined with precision the quantity 
 of chlorine required for the decomposition of a given amount of 
 water and ammonia ; the quantity of oxygen and nitrogen which 
 are evolved in each case respectively ; and the quantity of 
 hydrochloric acid produced in each case. These valuable 
 results may be concisely embodied in the following simple 
 equations : 
 
 1. Decomposition of water ~by chlorine, and products of the re- 
 action : 
 
 + C1 + C1 = HC1 + HC1 + 0; 
 
 or, more simply, 
 
 H 2 O + 2 Cl = 2HC1 + 0. 
 
 This equation not only illustrates the qualitative results of the 
 experiment, but giyes a comprehensive synoptical view of the 
 quantitative reaction which takes place, and of the weights of the 
 final products. To read this equation, we have only to attach to 
 each letter the name, weight, and volume, for which it stands, 
 when the proportions these bear to each other are at once made 
 manifest. Eeplacing the symbols by their ponderal values 
 H by 1, Q by 16, and Cl by 35'5, the expression becomes: 
 
 2 + 16 = 18 parts of water, require for their decomposition 
 2 x 35-5 = 71 parts of chlorine. 
 
90 DECOMPOSITION OF AMMONIA BY CHLORINE. 
 
 While the products are : 
 
 2 x (1 4- 35*5) = 73 parts of hydrochloric acid, and 16 parts 
 of oxygen. 
 
 Beading this equation, again, for the volume-ratios which it 
 also expresses, we learn from it that two volumes of water-gas 
 (the condensed product, as we already know, of the three 
 volumes H -f H + 0) require for their decomposition two 
 volumes of chlorine; the products formed being 2x2 = 4 
 volumes of hydrochloric acid, and 1 volume of free oxygen. 
 
 2. Decomposition of ammonia by chlorine, and products of the 
 reaction. 
 
 This decomposition, like the former one, may be expressed in 
 two ways ; we may either write : 
 
 H) 
 
 H I N + 01 + Cl + Cl = HC1 + HC1 + HC1 + N ; 
 HJ 
 
 or, more simply : 
 
 H 3 N + 3C1 = 3HC1 + N; 
 
 the interpretation of both expressions affording us precisely 
 similar information. 
 
 "Replacing the symbols by their numerical values, as before, 
 we obtain the following weight and volume ratios : 
 
 17 parts ammonia + 3 x 35*5 parts chlorine 
 
 2 vols. + 3 vols. 
 
 = 3 x (1 4- 35-5) parts hydrochloric acid -f- 14 parts nitrogen 
 
 6 vols. -f. 1 vol. 
 
 All this information, be it observed, is symbolized in the 
 preceding equation by eleven letters and figures, distributed into 
 four groups, joined by three relational signs. It is dimculLto 
 conceive a form of expression more concise and encyclopaedic. - 
 
 In the processes just reviewed we have had exclusively to deal 
 with elements which, at the common temperature, are gases ; and 
 
DECOMPOSITIONS EFFECTED BY THE ALKALI-METALS. 91 
 
 the volume-weights of which are, therefore, easy of determina- 
 tion. This gaseous form does not, however, characterize all 
 the elementary bodies, and hence arises a new and serious 
 obstacle in our path, when we endeavour to extend the 
 application of our newly-acquired language to solids. We 
 need not go beyond our present limited range of experience for 
 examples of this difficulty. It meets us even in the three 
 simple experiments which we selected to inaugurate our chemi- 
 cal studies. 
 
 It will be remembered that, in order to separate the hydrogen 
 from hydrochloric acid, from water, and from ammonia, we 
 submitted these compounds successively to the action of two 
 metallic elements, sodium and potassium, both of which are 
 solid bodies, not gases. The sum total of the information fur- 
 nished by these experiments, under the conditions stated, was the 
 fact of the evolution of hydrogen. The true nature of these re- 
 actions, the changes undergone by the sodium and potassium 
 employed, the weight-ratios in which the substances present act 
 upon one another, did not in those experiments press for imme- 
 diate consideration. These particulars come now in their turn 
 to be investigated, and embodied in appropriate symbolic ex- 
 pressions ; and, in setting ourselves to this task, we shall acquire 
 our first experience of the difficulty referred to above, as also of 
 the means by which it is to be overcome. 
 
 And first, as regards the qualitative character of these reactions. 
 With the experience acquired in studying the action of chlorine 
 upon water and ammonia, we can no longer have any doubt in 
 this respect. Sodium and potassium liberate the hydrogen from 
 hydrochloric acid, water, and ammonia, by withdrawing there- 
 from the elements chlorine, oxygen, and nitrogen, respectively 
 united therewith. The action of sodium upon hydrochloric 
 acid gives rise to the formation of a solid chlorine compound of 
 sodium, which we call chloride of sodium. The action of sodium 
 upon water, under appropriate circumstances, produces an 
 oxygen compound of sodium, also solid, termed oxide of sodium. 
 Lastly, the action of sodium, under favourable circumstances, 
 upon ammonia, gives rise to the formation of a third solid com- 
 
92 APPLICATION OF PONDEKAL ANALYSIS 
 
 pound of sodium ; that, namely, which it forms with nitrogen, 
 and which we may therefore call nitride of sodium. 
 
 But we have now, as before, to represent these processes in 
 equations, capable of disclosing the quantitative conditions of 
 the reactions. And here arises our difficulty. At this point we 
 meet with the obstacle opposed to our progress by the solidity, 
 at common temperatures, of one of the elements with which we 
 have to work. For, to accomplish our purpose to perfect our 
 knowledge of these reactions, and to embody our completed 
 knowledge in volumetric equations we evidently require to 
 know the gas-volume-weight of sodium; in other words, the 
 specific gravity of sodium-gas. Unfortunately, sodium can only 
 be volatilized at very high temperatures and under conditions 
 which make it extremely difficult to obtain the gas pure, and to 
 ascertain its volume-weight. Sodium-gas has never yet been 
 obtained in a state of perfect freedom from admixture, and its 
 volume-weight is consequently unknown to us. 
 
 In the absence of direct means of ascertaining the volume- 
 weight of this metal, Chemists have had recourse to an indirect 
 mode of research. They have striven to ascertain its combining 
 weight relatively to gaseous bodies ; and, for this purpose, they 
 have sought to obtain a gaseous combination of sodium with hy- 
 drogen. Could this be accomplished, the normal product-volume 
 of the compound gas (our well-remembered double unit-volume) 
 might be measured out and analyzed, and the weight of sodium 
 contained in it would represent the combining weight of this 
 body relatively to our standard, hydrogen. But this resource 
 also has failed. Chemists have hitherto been unable to produce 
 a hydrogen-compound of sodium. 
 
 Under these circumstances, nothing has remained but to fall 
 back upon ponderal analysis of the solid compounds of sodium 
 such as the chloride, the oxide, and the nitride of this body, 
 referred to in our previous experiments. With the value of 
 this expedient we are already acquainted. Ponderal analysis, 
 pure and simple, has already rendered us important service in 
 the study of hydrobromic and hydriodic acid, and of sulphuretted 
 and selenetted hydrogen ; while in the investigation of phospho- 
 
TO DETERMINE VOLUME-RATIOS. 93 
 
 retted and arsenetted hydrogen we had to rely almost exclu- 
 sively upon its indications. 
 
 Let us now see how far this method will serve our purpose in 
 the cases under consideration. 
 
 Weight-analyses of the three compounds generated by the 
 action of sodium upon hydrochloric acid, water, and ammonia, 
 have furnished the following results : 
 
 In chloride of sodium 
 
 35 -5 parts of chlorine (the weight of 1 vol.), are combined 
 with 23 parts of sodium. 
 
 In oxide of sodium 
 
 16 parts of oxygen (the weight of 1 vol.), are combined 
 with 23 x 2 = 46 parts of sodium. 
 
 In nitride of sodium 
 
 14 parts of nitrogen (the weight of 1 vol.), are combined 
 with 23 x 3 = 69 parts of sodium. 
 
 Equal volumes of chlorine, oxygen, and nitrogen, are thus 
 seen to combine with very unequal weights of sodium. The 
 weight of sodium combining with one volume of oxygen is 
 tioice, the weight combining with one volume of nitrogen is thrice, 
 the weight combining with one volume of chlorine. We thus see 
 that, in its behaviour towards chlorine, towards oxygen, and 
 towards nitrogen, sodium exactly resembles hydrogen ; for we 
 remember that one volume of chlorine, of oxygen, and of nitro- 
 gen, respectively combine with one, two, and three parts by 
 weight of hydrogen. 
 
 With these results before us, and bearing in mind the experi- 
 mental proofs already obtained that the weights of bromine 
 and of iodine respectively combining with one volume of hydro- 
 gen actually represent the gas-volume-weights of those elements, 
 are we justified in admitting the weight of sodium combined 
 with one volume of chlorine to be the volume-weight of sodium 
 gas? Further experiment is needed to answer this question. 
 There is, however, no valid objection to our provisional accept- 
 ance of this conception ; and, in this sense, we may adopt, as 
 
94 
 
 FOKMUL^I OF SODIUM-COMPOUNDS 
 
 the gas-volume weight of sodium, the value 23, representing 
 the weight of sodium which, in chloride of sodium, is united 
 with one volume of chlorine. The symbol of sodium is derived 
 from the Arabic name for soda, natron, whence natrium, the 
 German name for this metal ; the complete expression for this 
 element is accordingly Na = 23. 
 
 Thus much being taken for granted, the composition of the 
 three sodium compounds may be represented in formulaB, capable 
 of tersely and clearly setting forth their analogies with the 
 hydrogen compounds whose types of structure they respectively 
 affect. Thus, placing in one column the typical compounds, 
 and in the other their analogues, so as to bring them into com- 
 parison, side by side, we obtain the following series : 
 
 TYPES. 
 
 ANALOGUES. 
 
 Chloride of hydrogen 
 (hydrochloric acid) 
 HC1. 
 
 Oxide of hydrogen 
 (water) 
 
 1 ( O = H 2 
 
 Nitride of hydrogen 
 (ammonia) 
 H 
 
 H } N = H 3 N. 
 H 
 
 Chloride of sodium 
 (common salt) 
 NaCl. 
 
 Oxide of sodium 
 (soda) 
 
 Nitride of sodium 
 
 Na) 
 
 Na V N = Na 3 N. 
 Na 
 
 Whether the combining weight of sodium, represented by the 
 expression Na = 23, coincide with the relative volume-weight 
 of its gas ; or whether (as we found in the cases of phosphorus 
 and arsenic) the volume-weight of sodium be double its combin- 
 ing weight, so as to correspond to the expression 23 x 2 = 46 ; 
 or whether, lastly, these two values stand in some less simple 
 ratio to each other; these are questions which future experi- 
 ments can alone decide. 
 
CONTAINING CHLOKINE, OXYGEN, AND NITROGEN. 95 
 
 But, whatever may be the result of such ulterior investiga- 
 tions as to the gas-volume of sodium, we are enabled by our 
 present knowledge of its combining weight to represent, in the 
 following simple equations, the action of this metal upon hydro- 
 chloric acid, water, and ammonia, respectively : 
 
 HC1+ Na = NaCl + H 
 H S 4- 2Na = Na 2 + 2H 
 H 3 N + 3Na = Na 3 N + 3H 
 
 Whilst formerly we only knew sodium to be capable of libe- 
 rating hydrogen from its compounds with chlorine, oxygen, and 
 nitrogen, we now learn, from the above equations, that in order 
 to disengage one volume of hydrogen (H = 1 by weight) we 
 invariably require 23 parts by weight of sodium (Na) ; this 
 proportionality still holding good, whether the hydrogen com- 
 pound submitted to the action of the metal contain one, two, or 
 three volumes of hydrogen. We note this fact in passing ; its 
 full importance will become apparent to us a little further on in 
 our inquiry. 
 
 It is not, however, by sodium only that hydrochloric acid ? 
 water, and ammonia, are decomposed ; other metals act on them 
 in a similar manner. If, for example, potassium be substituted 
 for sodium in the foregoing experiments, hydrogen is equally 
 disengaged ; its evolution being of course attended by the forma- 
 tion of potassic instead of sodic chloride, oxide, and nitride. 
 Ponderal analysis of these compounds, coupled with a series of 
 considerations analogous to those we entered into in the case of 
 sodium, have led chemists to fix the combining weight of potas- 
 sium at 39 = K (the initial of Jcalium, a term of Arabic origin, 
 retained in the German as the name of the metal). The three 
 formulaB 
 
 TTl K ) 
 
 KC1, 1 0, and K I N, 
 
 KJ 
 
 represent these three potassium compounds respectively ; while 
 the equations 
 
96 PRESENT IMPERFECTION OF SYMBOLIC NOTATION. 
 
 HCl-f- K = KC1+ H 
 H 2 + 2K = K 2 + 2H 
 H 3 N -f 3K = K 3 N + 3H 
 
 illustrate the formation of these potassium compounds from the 
 corresponding hydrogen-compounds. These expressions may be 
 cited as further evidence of the terseness, lucidity, and precision, 
 introduced by the use of the language of symbols into the state- 
 ment and investigation of chemical transformations. 
 
 It is not, however, pretended that this system of chemical 
 notation has arrived at the perfect symmetry, and logical con- 
 sistency throughout, which we may hope it will attain with the 
 progress of chemical knowledge. At present, some of the rela- 
 tions which it is employed to express are assumptions based 
 only on highly-probable analogies. We have already seen, for 
 example, that, while the four symbols H, Cl, 0, and N, repre- 
 sent the volume-weights of elements which are gaseous at com- 
 mon temperatures and pressures, the four symbols Br, I, S, 
 and Se, refer to four elements which are not gases under 
 ordinary barometric and thermometric conditions, and whose 
 weights, under those conditions, can only be arrived at by 
 deduction from their observe.d volume-weights at the much 
 higher temperatures which do actually convert them into gases. 
 Again, while all the eight symbols above mentioned express the 
 weights of equal volumes of the several elements referred to, 
 the two symbols P and As express only the half-volume-weights 
 of the bodies which they are employed to denote. So, once 
 more, the symbols Na and K can be taken to represent the 
 unit-volumes of the alkaline metals they denote, only by pure 
 assumption ; seeing that experiment has hitherto failed to deter- 
 mine the respective volume-weights of those metals, when con- 
 verted into gas by intense heat. 
 
 In these instances, therefore, as well as in the cases of all 
 bodies (unfortunately, a large majority of the known elements) 
 which cannot be volatilized, or rendered gaseous, even by 
 the intensest temperatures as yet at our command, we are 
 
COINCIDENCE OF WEIGHT & VOLUME-RATIOS HOW FAR HYPOTHETIC. 97 
 
 compelled to rely exclusively upon ponderal analysis to deter- 
 mine their combining proportions, and from these to infer, 
 by aid of analogy, their probable volume-weights, assuming 
 them to be susceptible of volatilization. In all these instances 
 we are liable to be misled by analogy; to assume, for ex- 
 ample, coincidence of the combining weight with the volume- 
 weight of the body in question (as observed in the cases 
 of bromine and iodine, of sulphur and selenium) ; whereas, in 
 truth, there may be, as in the cases of phosphorus and arsenic, 
 no such coincidence, but an exceptional divergence ; the com- 
 bining weight corresponding to the half volume only, not to 
 the full volume of the gas ; or even, perhaps, standing in some 
 less simple ratio thereto. 
 
 It is our duty to bear in mind that, in this, as in all other 
 respects, and in our own as in all antecedent ages, an imperfect 
 and transitional condition still characterizes scientific progress, 
 and impresses a correspondingly uncertain and provisional 
 character upon our best efforts to interpret and extend it. 
 
 It is only, therefore, in this limited sense, and subject to 
 these philosophical reservations, that the assumed coincidence 
 of the volumetric with the ponderal structure of chemical com- 
 pounds is advanced in those cases which, for the present, lie 
 beyond the range of experimental verification. 
 
98 
 
 LECTUKE VI. 
 
 Fourth term in the series of typical hydrogen-compoundsmarsh-gas, or 
 light carbonetted hydrogen reasons for its separate consideration its 
 occurrence in marshes in coal-mines in coal-gas its preparation its 
 distinctive characters its qualitative analysis separation therefrom 
 of carbon by chlorine. Decomposition of marsh-gas by heat - its quan- 
 titative analysis its synthesis not yet accomplished combining weight 
 of its constituent, carbon its analogues. Silicetted hydrogen proba- 
 bility of the construction thereof upon the type of marsh-gas. 
 
 IN the series of typical hydrogen compounds which we have 
 now studied, we have found one volume each of chlorine, oxygen, 
 and nitrogen, united respectively with one, two, and three 
 volumes of hydrogen ; the condensation increasing in direct 
 ratio with the increasing proportion of hydrogen, so that, in 
 each case, two volumes only of the resultant compound are 
 produced. 
 
 This typical series does not end here, however. It comprises 
 a fourth member a compound containing, in two volumes, four 
 volumes of hydrogen condensed ; the other constituent being 
 the body familiar to all as carbon. This compound is the light, 
 inflammable gas, too well known to coal-miners as Fire-damp ; 
 and also termed, on account of its frequent emanation from 
 boggy ground, Marsh-gas. 
 
 The description of this fourth typical hydrogen-compound 
 might have been included with that of the rest of the series, 
 but we have preferred to reserve it as a matter for separate study, 
 chiefly on account of this signal difference ; that whereas in all 
 the other members of the series both constituents of the com- 
 pound are gaseous, in this fourth one the hydrogen is united 
 with a body which is not only solid at common temperatures, 
 but incapable of volatilization by any, the intensest means at 
 our present command. Hence, while the first three compounds, 
 studied collectively, apart from the fourth, illustrated, with an 
 
MARSH-GAS. 99 
 
 admirable symmetry, and in an unbroken ascending scale, the 
 laws of combination and progressive condensation, by volume 
 as well as by weight, the fourth exemplifies volume com- 
 bination and condensation only so far as its gaseous ingre- 
 dient is concerned ; our positive knowledge, as to its solid con- 
 stituent, extending only to the iveight-T&tio, and any views we 
 may entertain as to the volume-T&tio of this element being of 
 necessity speculative. It is, indeed, maintained by many che- 
 mists, and not without some show of reason, that analogy affords 
 a fair basis for such speculation; but not until carbon shall 
 have been actually volatilized, and its vapour actually weighed, 
 shall we be entitled to rank marsh-gas, with respect to its 
 volumetric constitution, on the same certain footing as experi- 
 ment already assigns to hydrochloric acid, water, and ammonia. 
 
 In describing this as the chief, we imply that it is not the only 
 difference which justifies us in separating the study of marsh- 
 gas from that of the three other typical compounds. A pecu- 
 liarity in its chemical deportment also distinguishes marsh-gas 
 from the other members of the typical series, in a manner which 
 will call for our special attention hereafter, but of which it 
 would be premature to say more, at present, than that its 
 effect is to prevent our employing, in the case of marsh-gas, 
 certain modes of investigation, which we have adopted with 
 advantage in studying the other typical compounds. 
 
 Thus much premised, let us proceed to make acquaintance 
 with this fourth member of our typical series. 
 
 There escapes from the fissures of the great coal-measures a 
 transparent, colourless, inflammable gas, which frequently accu- 
 mulates in the galleries of ill-ventilated coal-mines ; and, when 
 ignited by the miner's candle, through neglect of the protective 
 precautions provided by science, gives rise to the explosions so 
 much dreaded and deplored on account of their life-destroying 
 violence. Many varieties of coal so abound in this gas, that it may 
 be seen rising therefrom in bubbles, when newly-dug fragments 
 are thrown into water. The same gas, as we have already men- 
 
 r 2 
 
100 SOURCES OF MARSH-GAS. 
 
 tioned, is developed in marshy lands, and is often observed 
 bubbling up from stagnant pools and swamps, where vegetal 
 matters are in process of gradual decay. In summer, when decay 
 is most active, the development of this gas is most abundant ; and 
 it may be readily collected in a glass cylinder, inverted over, 
 and plunging into, stagnant waters. The gas thus obtained may 
 be at once distinguished from common air by applying a light, 
 when it is found to be inflammable. 
 
 It would, of course, be inconvenient to procure this com- 
 pound from either of its natural sources, even if they fur- 
 nished it in a state of purity, instead of, as always happens, 
 yielding it in admixture with common air and other gases. The 
 ordinary illuminating gas, distilled from coal by artificial heat, 
 always contains a large percentage of marsh-gas ; but here 
 also it is mixed with other gases, scarcely separable from it ; so 
 that we are debarred from this source of supply, otherwise so 
 accessible and abundant. Chemists have, however, devised a 
 simple process by which marsh-gas may be readily prepared in 
 any quantity requisite for laboratory purposes, from well-known 
 materials, procurable at relatively moderate cost. In a flask (of 
 glass, or better, of copper or iron), arranged for gas disengage- 
 ment, strong vinegar is heated with a mixture of lime and the 
 caustic soda of commerce ; after a short time a transparent colour- 
 less gas is evolved, which is collected over water in the usual 
 way (Fig. 60). We have, for the present, no interest in dwelling 
 on the reaction which, under these circumstances, gives rise to 
 marsh-gas. Suffice it to say that a portion of the carbon con- 
 tained in the vinegar combines with the hydrogen present, so 
 as to form this gas ; which is also frequently called light car- 
 buretted or carbonetted hydrogen. For brevity's sake, how- 
 ever, we will, on the present occasion, retain the name of 
 marsh-gas. 
 
 From the hydrogen compounds previously examined, marsh- 
 gas is readily distinguishable by its inflammability. Kindled 
 at a taper it burns with a feebly-luminous flame. Indepen- 
 dently of this character, however, marsh-gas is strikingly dis- 
 
DISTINCTIVE CHARACTERS OF MARSH-GAS. 
 
 101 
 
 tingtiished from hydrochloric acid and ammonia gases, by its 
 possessing neither odour nor action on vegetal colours ; which 
 negative characters it shares with water-gas. 
 
 FIG. 60. 
 
 Not less easily may marsh-gas be distinguished from the three 
 elementary gases which we have submitted to examination. In- 
 deed, this readily inflammable gas cannot possibly be confounded 
 either with chlorine, oxygen, or nitrogen, which are all incom- 
 bustible bodies ; the two former being supporters of combustion, 
 but themselves not inflammable, while the last-named neither 
 burns, itself, nor sustains the combustion of other bodies. From 
 chlorine, moreover, marsh-gas differs by its lack of colour, 
 odour, and bleaching property; from oxygen, marsh-gas, like 
 nitrogen, is distinguishable by its total incapacity for support- 
 ing the process of combustion. The only elementary gas for 
 which, at the first glance, the hydrogen compound of carbon is 
 at all likely to be mistaken, is hydrogen itself; seeing that 
 inflammability, coupled with the absence of colour, of odour, 
 of bleaching power, and of the property of supporting combus- 
 tion, characterise both these gases. The difference, however, of 
 
102 
 
 ACTION OF CHLORINE ON MARSH-GAS. 
 
 FIG. 61. 
 
 hydrogen from carbonetted hydrogen becomes at once obvious 
 if both gases be burned side by side : for, while hydrogen burns 
 with a non-luminous and scarcely-visible flame, the combustion 
 of marsh-gas is attended by the evolution of feeble, but unmis- 
 takeable rays of light. 
 
 The chemical difference of the two gases may be demonstrated 
 by a simple experiment. We remember that a mixture of hydro- 
 gen and chlorine, when ignited, merely gave rise to the forma- 
 tion of hydrochloric acid. A similar experiment performed with 
 a mixture of marsh- gas and chlorine produces, as we shall see. 
 a further and very striking result. 
 
 For this purpose, we fill a tall cylinder with warm water, 
 invert it over the pneumatic trough, and pass marsh-gas into it 
 until a little more than one-third of the water is displaced ; 
 which done, we fill the two remaining thirds of the cylinder with 
 
 chlorine, taking care to pro- 
 tect the vessel, during manipu- 
 lation, from exposure to sun- 
 light. The cylinder thus filled 
 we close, by slipping a glass 
 plate beneath its mouth ; then, 
 raising it from the trough, 
 we agitate it to mix the gases 
 thoroughly ; and finally, we 
 apply a light to the mixture. 
 Ignition takes place, with pro- 
 duction of hydrochloric acid, 
 by the union of the chlorine 
 with the hydrogen present, as 
 in our former experiment with 
 pure hydrogen and chlorine. 
 But the marsh-gas betrays its 
 additional constituent, by a 
 copious separation of carbon 
 
 as the flame descends into the cylinder; the sides of which 
 become coated with a dense black layer of soot (Fig. 61). 
 
COMPOSITION OF MARSH-GAS. 103 
 
 The interpretation of this phenomenon presents no difficulty. 
 Chlorine acts upon marsh-gas as, under appropriate conditions, 
 we have scon it acting upon water and ammonia. Water-gas, 
 under the influence of chlorine, yields its oxygen, and ammonia 
 its nitrogen; in a similar manner, marsh-gas gives off its 
 carbon. The hydrogen withdrawn, in all three cases, combines, 
 in the very act of separation, with chlorine, to form hydrochloric 
 acid gas. 
 
 The action of chlorine upon marsh-gas indubitably points to 
 hydrogen and carbon as elementary constituents of this com- 
 pound ; and it therefore remains only to demonstrate, by syn- 
 thesis, that hydrogen and carbon are its exclusive constituents. 
 But here we meet with difficulties very similar to those which 
 compelled us to forego the synthesis of ammonia. No means 
 are known at present by which hydrogen and carbon can be 
 directly combined to form marsh-gas. Nevertheless, in this 
 case, as in that of ammonia, the balance comes to our aid, and 
 enables us to decide the question. The sum of the weights of 
 the two constituents separated from a given quantity of marsh - 
 gas is found to be precisely equal to the weight of the marsh- 
 gas submitted to experiment. 
 
 Our next task is to determine the combining weight of the 
 carbon ; and, for this purpose, to investigate the ratio in which 
 this body is united with hydrogen in marsh-gas. 
 
 Were carbon a gaseous element, we should, of course, have 
 no difficulty in ascertaining its combining weight, by the volu- 
 metric method employed in the cases of chlorine, oxygen, and 
 nitrogen. Were carbon even susceptible of volatilization, we 
 might endeavour to determine its volume-weight directly by aid 
 of the balance ; though here we should be liable to meet with 
 just such a disparity between the combining and volume weights 
 as exceptionally obtains in the case of phosphorus and arsenic. 
 But, in point of fact, as we have already seen, carbon is a solid 
 which it is not as yet in our power to convert into gas : and 
 therefore we are constrained to fall back, in its case, upon the 
 only universally applicable method with which we have thus far 
 
104 
 
 VOLUME-WEIGHT OF MARSH-GAS. 
 
 become acquainted, for ascertaining, the combining weights of 
 elementary bodies viz., the determination of the weight thereof 
 contained in the product-volume of that one of its hydrogen- 
 compounds in which it is present in the smallest quantity. 
 
 Accordingly, we represent by C the quantity of carbon 
 contained in 2 vols. of marsh-gas; and this we admit as the 
 combining weight of carbon. The task before us resolves 
 itself, therefore, into the exact determination, by means of the 
 balance, of the quantities of hydrogen and carbon respectively 
 
 FIG. 62. 
 
 contained in 2 volumes of marsh-gas. The solution of this pro- 
 blem, as we shall hereafter see, is furnished by a single experi- 
 ment, not less elegant than conclusive. Meanwhile we must be 
 contented to employ a somewhat circuitous and less rigorous 
 method, lying within the scope of our present chemical expe- 
 rience. 
 
 For this purpose, our first care is to ascertain the volume- 
 weight of marsh-gas. That this gas is much lighter than atmo- 
 spheric air is readily seen by simply filling a cylinder with 
 marsh-gas, allowing it to stand uncovered for a few minutes, 
 and then applying a lighted taper. Ignition does not ensue ; 
 every trace of the inflammable gas is found to have escaped ; 
 the heavier common air having taken its place in the vessel. 
 Careful experiments have proved that marsh-gas is exactly 8 
 times heavier than hydrogen ; hence, taking the unit- volume of 
 hydrogen as 1, two volumes of marsh-gas weigh 16. 
 
 How much hydrogen is present in these 2 volumes ? This 
 question admits of being answered, in very close approximation, 
 
ITS EXPERIMENTAL DETERMINATION. 
 
 105 
 
 
 by the method which served us in the case of ammonia, viz., by 
 splitting it up into its constituents under the influence of 
 heat. This experiment we perform, as before, in a U tube, 
 fitted up with spark wires (Fig. 62). As our source of heat, 
 KI.;. 6 >. we again employ the spark- 
 
 current of the induction-coil. 
 No sooner does the spark-cur- 
 rent begin to traverse the tube, 
 than the marsh-gas is found to 
 expand ; and, after the lapse of 
 a few minutes, a light deposit 
 of carbon is formed in the 
 vicinity of the platinum wires. 
 The decomposition, energetic 
 at the commencement of the ex- 
 periment, proceeds more slowly 
 as the gas dilates; so that a 
 considerable time is required 
 
 to bring the operation to a close. If the mercury be allowed 
 to run out from the nipper-tap, till it has become level in the 
 two limbs of the U tube, it is found that the original gas-volume 
 has very nearly doubled. This result having been attained, the 
 continued transmission of the spark-current produces no further 
 increase in the volume of the gas ; which is then readily shown, 
 by the usual tests, to have lost the characteristic properties of 
 marsh-gas ; to be, in fact, pure hydrogen. 
 
 This experiment presents greater difficulty of manipulation 
 than attends the processes we employed for determining the com- 
 position of hydrochloric acid, water, and ammonia. The spark- 
 current, when transmitted through marsh-gas, is not unfrequently 
 interrupted by the formation of a conducting bridge of carbon 
 between the platinum points ; and this conductor has, of course, 
 to be destroyed, in order that the current may have an interval 
 through which to leap in the form of sparks. In order to break 
 the bridge, we may agitate the mercury in the tube until it 
 reaches the platinum wires. Or, better still, we may endeavour 
 
 F 3 
 
106 
 
 DIAGRAMMATIC EQUATION OF MARSH- GAS. 
 
 to prevent the formation of the carbon-bridge by reversing from 
 time to time the direction of the current. 
 
 With these precautions, the experiment affords very closely 
 approximative, though not perfectly exact, results. Slight error 
 arises from the circumstance that a minute proportion of marsh- 
 gas escapes decomposition by the spark-current ; and, accord- 
 ingly, the volume of hydrogen obtained is never quite double 
 the volume of the marsh-gas submitted to the treatment. Never- 
 theless, the experiment, even in its present imperfect form, 
 unequivocally points to a result, which hereafter, by other 
 methods, we shall establish beyond question; namely, that 
 marsh- gas contains twice its volume of hydrogen. 
 
 Accordingly, two volumes of marsh-gas, the weight of which 
 we have found to be 16, contain 4 volumes (and also, there- 
 fore, 4 parts by weight) of hydrogen, combined with 16 4 = 
 12 parts, by weight, of carbon. This quantity (C = 12) is, 
 therefore, the combining weight of carbon, i. e., the weight of 
 carbon contained in two volumes of marsh-gas. We are now, 
 therefore, justified in representing the composition of marsh-gas 
 by the diagrammatic equation 
 
 > + i C j 
 
 Dispensing with the squares, we obtain the short formula 
 4H + C = H 4 C, 
 
 the second term of which represents the weight of two volumes 
 of this gas, exactly as the expressions HC1, H 2 0, and H 3 N express 
 
ITS TYPICAL CHARACTER. 107 
 
 the weights of two volumes of hydrochloric acid, water-gas, 
 and ammonia. 
 
 It will be observed that, in the diagram, we have surrounded 
 the element C with a square formed in dotted, instead of full 
 lines ; an expedient which we shall find convenient on future 
 occasions, as well as in the present case, to signify that the 
 volume represented is hypothetical.- The full lines will always 
 be reserved to symbolize volumes placed beyond question by 
 actual experiment. 
 
 In marsh-gas we have added a new term to the series of hy- 
 drogen-compounds which have so long engaged our attention. 
 Hydrogen does not, so far as we yet know, form with any element 
 whatsoever any compound, whereof the normal product-volume 
 contains more than four volumes of hydrogen united with the 
 combining weight of the other element. Marsh-gas may, there- 
 fore, be considered as an example of the hydrogen-compounds 
 richest in hydrogen ; exactly as we regard hydrochloric acid as 
 exemplifying the hydrogen-compounds poorest in hydrogen ; and 
 water and ammonia as representing hydrogen-compounds of 
 intermediate composition. These relations are clearly displayed 
 in the following series of formulae : 
 
 Hydrochloric acid . . H Cl = 2 vols. 
 
 Water H 2 = 2 
 
 Ammonia H 3 N = 2 
 
 Marsh-gas .. .. .. H 4 C =2 
 
 It remains to justify the introduction of marsh-gas into the 
 series of typical hydrogen-compounds ; in other words, to show- 
 that it deserves to be regarded as the model upon which one 
 or more analogous compounds are formed. 
 
 In order definitively to raise marsh-gas to this rank, we ought, 
 evidently, to be in possession of some binary hydrogen-com- 
 pound or compounds, each containing some constituent analogous 
 to carbon, and, moreover, containing it in such relative propor- 
 tions that, with its minimum combining weight there may be 
 
108 
 
 ANALOGUES OT MARSH-GAS. 
 
 united four volumes of hydrogen ; the whole being reduced, by 
 condensation, to one normal product-volume. 
 
 Only one such analogue of marsh-gas has hitherto been ac- 
 tually obtained ; and even this is but a recently discovered and 
 as yet imperfectly studied compound. It is the compound of 
 hydrogen with silicon ; and so far as experiment has yet made 
 out its composition, this body appears to contain 
 
 + i Si ; 
 
 i 1 
 
 H 4 Si 
 
 Or, dropping the squares as before, 
 
 4H + Si - H 4 Si ; 
 
 in which Si ( = 28*5) represents the combining weight of silicon. 
 That silicon does really thus combine with hydrogen in the 
 ratio above represented, is rendered all the more probable by the 
 fact that, in forming compounds with other elements, carbon and 
 silicon resemble each other as to the proportions of such ele- 
 ments which they respectively take up. Analogies of a similar 
 kind justify the anticipation that hydrogen-compounds of titanium 
 and tin will be obtained, having the same proportional composi- 
 tion as the hydrogen-compound of silicon, and forming, with it, a 
 normal series of hydrogen-compounds grouped under the type of 
 marsh-gas. These elementary bodies have not yet been volati- 
 lifced, it is true ; but so neither has carbon itself. This circum- 
 stance, therefore, will only oblige us to adopt, as their combining 
 weights, so much of each body as the balance may prove to exist 
 in two volumes of the hydrogen-compound it may be found to 
 
RESULTS OF THE INQUIRY THUS FAR. 109 
 
 produce ; guarding ourselves against the symbolic affirmation 
 of more than we actually know, by representing, as before, the 
 assumed values by dotted squares. 
 
 From the series of phenomena which we have now studied 
 together, we have already begun to deduce some valuable general 
 conceptions. By the experimental examination of a small 
 number of bodies we have learned to contradistinguish ele- 
 mentary bodies from compounds, and mechanical mixture from 
 chemical combination. We have also been led, step by step, to 
 recognize the conditions of immutable proportionality under 
 which chemical compounds are formed, and the simple volume- 
 ratios in which the elements associate for their production. 
 The comprehension of these ratios induced us naturally to seek 
 a concise yet exact language for their expression, and this we 
 found in the symbolic system of chemical notation. Eeferring 
 to experiments, few but conclusive, for the proof of our posi- 
 tions, and the guidance of our progress, we were enabled to 
 examine, by their certain light, results gathered in a much wider 
 sphere than we have yet mastered for ourselves ; and thus, by 
 the alternate employment of the inductive and the deductive 
 methods, we have been enabled to construct a general system, in 
 which many particular facts may be embraced. 
 
 This is something to have achieved, but much more remains 
 to be done. Let us bear in mind that our notions rest as yet 
 upon the study of a very limited number of bodies, elementary 
 and compound ; let us be prepared for considerable extension, 
 and at the same time clearer definition of our views, as our 
 knowledge of the facts to be co-ordinated acquires a wider 
 range. 
 
 Meanwhile we have still much to learn without going beyond 
 the five elements, hydrogen, chlorine, oxygen, nitrogen, and 
 carbon, to which we have confined our attention. Thus far we 
 have examined only the typical compounds formed by hydrogen 
 with the four other elements just enumerated ; and these typical 
 hydrogen compounds are doubtless of paramount importance as 
 
110 VAKIED COMBINATIONS OF THE TYPICAL ELEMENTS. 
 
 forming the main pillars of our edifice. Nevertheless, we shall 
 have hereafter to make acquaintance with a countless variety 
 of compound bodies, in whose composition hydrogen bears no 
 part ; and these are well exemplified in the varied combinations 
 formed amongst each other by chlorine, oxygen, nitrogen, and 
 carbon. The exhaustive study, even of this narrow field, would 
 go far to prepare us for mastering the much wider domain that 
 lies beyond ; but for the purposes of the present introduction 
 necessarily restricted in its scope we will select from among 
 these manifold subjects one only for immediate consideration 
 that, namely, which relates to the deportment of nitrogen towards 
 oxygen. 
 
Ill 
 
 LECTURE VII. 
 
 Deportment of nitrogen towards oxygen nitric acid hydrated anhy- 
 drous its composition its decomposition by heat by metals by tin, 
 yielding hyponitric acid by silver, yielding nitrous acid by copper, 
 yielding nitric oxide by zinc, yielding nitrous oxide characters of 
 these products how shown to be chemical compounds, not mechanical 
 mixtures expansion of the idea of chemical combination combination 
 of two elements in several proportions law of multiple proportions 
 volume and condensation ratios in chemical compounds ordinary 
 exceptional. 
 
 HAVING selected as the next subject of our inquiry the deport- 
 ment of nitrogen towards oxygen gas, we are led at once to the 
 study of one of the most intense and interesting agents existing 
 in the whole range of chemistry namely, Nitric acid. 
 
 Nitric acid is a powerfully-corrosive liquid, long known by 
 the name AQUAFORTIS, and now manufactured on a large 
 scale for industrial purposes. This liquid contains, together 
 with water, a compound of nitrogen and oxygen, which, when 
 entirely separated from water, is called anhydrous nitric acid, 
 or nitric anhydride. This separation is not, however, so easily 
 effected as in the case of hydrochloric acid ; the aqueous solu- 
 tion of which, as you remember, only required heating to ebulli- 
 tion for this purpose. 
 
 By appropriately selected means, the examination of which 
 we must defer, the dehydration of nitric acid has been neverthe- 
 less accomplished. 
 
 Anhydrous nitric acid presents itself in the form of white 
 fusible crystals, in which analysis has proved the presence of 
 two volumes of nitrogen combined with five volumes of oxygen, 
 or 28 parts by weight of the former with 80 parts of the latter 
 element. Anhydrous nitric acid is a somewhat instable com- 
 pound. It is decomposed even by gentle heat, with evolution 
 
112 
 
 NITRIC ACID ITS DECOMPOSITION BY HEAT, 
 
 of copious red fumes, which, when passed through water, are 
 for the most part absorbed, the unabsorbed portion being a 
 colourless gas, which is readily recognised as oxygen. 
 
 It is not, however, in its anhydrous condition only that nitric 
 acid is thus easy of decomposition. The ordinary hydrated 
 nitric acid yields, on ebullition, vapours which, when heated to 
 dull redness, are decomposed into a similar mixture of red 
 fumes and colourless oxygen. Fig. 63 shows an apparatus 
 
 for performing this experiment. The nitric acid is boiled in the 
 sealed and down-bent extremity of a tube, the continuation of 
 which is kept red-hot by gas flames, so as to effect the decompo- 
 sition of the vapours evolved. The same tube, further pro- 
 longed, conducts the gaseous products of decomposition to the 
 trough, for collection of the oxygen over water; the elbow in 
 this conducting portion of the tube serving to retain the acid 
 liquors of condensation. In this experiment a compound, con- 
 sisting of two volumes of nitrogen and five volumes of oxygen, 
 abandons a certain proportion of its oxygen ; and it is therefore 
 obvious that, whatever may be in other respects the nature of 
 the red fumes absorbable by water, they must at all events con- 
 tain less than five volumes of oxygen in association with two of 
 nitrogen. 
 
 The separation of this red fuming product from nitric acid 
 
BY TIN, SILVER, COPPER, ZINC. 113 
 
 may be accomplished, even at ordinary atmospheric tempera- 
 tures, by submitting the acid to the action of bodies possess- 
 ing, like most of the metals, considerable attraction for oxygen. 
 Tin, silver, copper, zinc, when immersed in nitric acid are 
 powerfully attacked, with disengagement of gaseous products 
 as before, but without escape of oxygen this being appro- 
 priated by the metals which it converts into oxygen compounds 
 termed oxides. The quantity of oxygen thus withdrawn from 
 nitric acid varies with the nature of the metal used, and with 
 the physical conditions of the experiment, especially with the 
 temperature at which the operation is conducted, and the dilute- 
 ness of the acid employed. 
 
 Tin, made to act on nitric acid under favourable circum- 
 stances, removes one-fifth of its oxygen with disengagement of 
 a volatile compound, which at a low temperature condenses to 
 white needles. This compound, which is termed hyponitric acid, 
 contains two volumes of nitrogen united with four volumes of 
 oxygen ; or by weight, 28 of the former combined with 64 of 
 the latter element. 
 
 When submitted under appropriate conditions to the action of 
 silver, nitric acid loses two-fifths of its oxygen, and is converted 
 into a yellowish-red vapour, which, on cooling, condenses into a 
 bluish-green liquid. This substance is called nitrous acid, and 
 contains two volumes of nitrogen combined with three of oxygen, 
 or by weight 28 of the former to 48 of the latter. 
 
 Copper, when brought into contact with nitric acid, appro- 
 priates three-fifths of its oxygen with evolution of a transparent 
 colourless gas, which consists of two volumes of nitrogen and 
 an equal bulk of oxygen gas ; or, by weight, 28 to 32 parts of 
 the respective elements. This product, which is termed nitric 
 oxide, has the remarkable property of acquiring a yellowish- 
 brown colour by exposure to the air. 
 
 Under the influence of zinc, lastly, if the experiment be well 
 conducted, nitric acid is deprived of not less than four-fifths of 
 its oxygen, with evolution of a colourless gas, somewhat soluble 
 in water, and which also remains colourless in contact with the 
 
114 
 
 FIVE COMPOUNDS OF NITROGEN WITH OXYGEN. 
 
 air. This gas, which is called nitrous oxide, contains two volumes 
 of nitrogen associated with one volume of oxygen, or by weight 
 28 parts of the former to 16 of the latter gas. 
 
 Thus, then, there are no less than five compounds of nitro- 
 gen with oxygen, the composition of which by volume and by 
 weight, with the corresponding formulae, is set forth in the fol- 
 lowing table : 
 
 NAMES. COMPOSITION. FORMULA. 
 
 By volume. 
 Nitrogen. Oxygen. 
 
 Nitrous oxide .. 2 vols.+l vol. 
 Nitric oxide . . 2 vols. -j- 2 vols. 
 Nitrous acid . . 2 vols. -f- 3 vols. 
 Hyponitric acid 2 vols. -f- 4 vols. 
 Nitric acid . . 2 vols. -j- 5 vols. 
 
 By weight. 
 Nitrogen. Oxygen. 
 
 14x2 = 28+ 16 
 
 28 + 16x2 = 32 
 
 28 + 16x3 = 48 
 
 28 + 16x4 = 64 
 
 28 + 16x5 = 80 
 
 N 2 
 
 NA 
 NA 
 NA 
 
 NA 
 
 The first thing that strikes us, on considering with atten- 
 tion this remarkable series of compounds, is the contrast they 
 present, by their multiplicity, to the products obtained in all 
 our previous experiments. As yet we have become acquainted 
 with only one chemical compound of hydrogen with chlorine, 
 of hydrogen with oxygen, and of hydrogen with nitrogen, 
 though we could obtain mechanical mixtures of those gases 
 in any proportions. We remained, therefore, within the limits 
 of our then acquired knowledge, in considering chemical com- 
 bination to be contradistinguished from mechanical mixture, 
 by the single and immutable proportionality we had always 
 seen the former affect, while the latter admitted manifold pro- 
 portions. 
 
 In the presence of the phenomena we have just witnessed, this 
 conception can evidently no longer be maintained. 
 
 If the five bodies just examined be chemical compounds, not 
 mechanical mixtures, our former view of chemical combination 
 must be modified and expanded to include them. And that they 
 are true chemical compounds we have unequivocal proofs ; first, 
 in the definiteness and constancy of their composition ; secondly, 
 
COMBINATION IN MULTIPLE PROPORTIONS. 115 
 
 in the essential differences of property and character that dis- 
 tinguish them from their elementary constituents. Nitrogen 
 and oxygen are colourless gases, insoluble in water, and in- 
 capable of condensation into the liquid (and a fortiori into the 
 solid) form by any means at our command. Nitric anhydride 
 and hyponitric acid, on the other hand, solidify at low tem- 
 peratures to white crystalline bodies. Nitrous acid condenses 
 by cold to a yellow-brown liquid; nitric oxide in contact 
 with air acquires a bluish-green colour ; nitrous oxide, lastly, 
 is much more soluble in water than either of its con- 
 stituents. It is therefore obvious that these bodies are not 
 mere mechanical mixtures, but true chemical compounds of 
 oxygen and nitrogen ; and we are thus irresistibly led to the 
 conclusion one evidently of the highest importance that two 
 of the elements are capable of combining in several proportions 
 to form a series of true chemical compounds, each differing from 
 the others, and all differing from their primary constituents. 
 
 The difference between a mechanical mixture and a chemical 
 compound does not, however, on this account become less 
 sharply marked. In mechanical mixtures the elements may 
 associate in proportions, whose name is legion, and which may 
 be multiplied to any extent by arbitrary increments of this or 
 that ingredient in the mass. In chemical compounds, on the 
 contrary, the elements are united in comparatively few propor- 
 tions ; those which succeed to the first being all referrible thereto, 
 as its multiples, in definite ratios. On the one hand limited, on 
 the other boundless variety ; and on the one hand well-defined, 
 on the other indeterminate proportions. The possible mixtures 
 of nitrogen with oxygen are beyond the power of numbers to ex- 
 press ; the possible combinations of nitrogen with oxygen are only 
 five ; two volumes of nitrogen associating chemically, as we have 
 seen, either with one, two, three, four, or five volumes of oxygen, 
 and with no other proportions whatsoever. And should we ever 
 succeed in producing other true compounds of nitrogen with 
 oxygen, all chemical experience assures us that such combina- 
 tions will take place in equally determinate ratios. 
 
116 LAW OF MULTIPLE PROPORTIONS. 
 
 And herein, again, our study of the compounds of nitrogen with 
 oxygen affords us a further lesson of the deepest interest. Not 
 only is the number of these compounds limited, but the indi- 
 vidual compounds, as far as their composition is concerned, bear 
 to each other relations of the utmost simplicity. This simplicity 
 is conspicuously displayed in the volume column of our table 
 of the nitroxygen compounds. The volumetric proportion of 
 oxygen united with two volumes of nitrogen is observed to 
 advance in multiple proportions ; the second, third, fourth, 
 and fifth compounds containing respectively a double, treble, 
 quadruple, and quintuple bulk of oxygen, as compared with the 
 volume of this element in the first. There is nothing indetermi- 
 nate in this series, as in a series of mechanical mixtures. The 
 nitrogen volume being taken as constant in all the compounds, 
 the oxygen volumes rise by definite bounds ; and, as with 
 volumes, so of course with weights. In the second column of 
 the table we see the rise of the oxygen weight per saltum, in 
 multiple ratios, throughout the series ; and these definite pro- 
 portions, both of volume and weight, are jointly expressed in 
 the formulas of the last column. 
 
 These facts, though gathered from the study of a single series 
 of compounds only, illustrate one of the most general and com- 
 prehensive laws of chemistry ; the master-principle which, as 
 we shall hereafter find, connects and co-ordinates its simplest 
 and most elementary truths with their furthest and most intri- 
 cate developments. 
 
 The combinations of chlorine with oxygen form a series analo- 
 gous to that of the compounds of nitrogen with oxygen just 
 examined ; while hydrogen and oxygen, which as yet we have 
 seen associated only in water, form a second compound, called 
 peroxide of hydrogen, in which a given amount of hydrogen is 
 united with twice as much oxygen as is therewith associated in 
 water. 
 
 These observations, however, by no means justify us in assum- 
 ing that each couple of elements necessarily combines in more 
 than one proportion. On the contrary, hydrogen, so far as our 
 
SIMPLEST FORMULAE OF THE NITROXYGEN SERIES. 117 
 
 present information goes, combines with chlorine in one propor- 
 tion only ; that, namely, which produces hydrochloric acid. And 
 so, again, hydrogen forms with nitrogen no other combination than 
 that which results in the production of ammonia. 
 
 The composition of the five nitroxygen compounds, which 
 has already so largely expanded our knowledge, still claims 
 our brief attention from one other point of view. 
 
 The assumption of a constant amount (by weight or volume) 
 of nitrogen in all these compounds is well calculated to exhibit 
 the unbroken regularity of the steps by which the proportion of 
 oxygen advances from the beginning to the end of the series. 
 But this mode of representation is, on the other hand, disad- 
 vantageous in so far as it does not assign the simplest possible 
 expression for two compounds of the series. On glancing at the 
 column of formulae in the table (repeated here for convenience) 
 
 N,O NA NA NA NA, 
 
 we perceive that for the second and fourth formulae, N 2 (nitric- 
 oxide), and NA (hyponitric acid), the simplest expressions 
 would be obtained by halving each respectively : thus 
 
 is the simplest expression for nitric oxide ; and 
 
 is the simplest expression for hyponitric acid. Thus corrected, 
 the series of formulae representing the nitroxygen series assumes 
 this shape : 
 
 N 2 NO NA N0 2 NA- 
 Nor is their simplicity the only advantage resulting from this 
 modification of the two formulae in question. On inquiring into 
 the product-volume thus expressed, in each of these two cases, we 
 find that it coincides exactly with the product-volume expressed by 
 the formula N 2 representing the first term of the nitroxygen 
 series ; and we find, further, that this product-volume coincides 
 
118 VOLUMETRIC SYMBOLIZATION 
 
 exactly with our normal product-volume, as determined in the 
 typical cases of hydrochloric acid (HC1), of water (H 2 0), of am- 
 monia (H d N), and of marsh-gas (H 4 C). The product-volume we 
 remember was, in all these cases, double the hydrogen unit- 
 volume. This normal product-volume careful experiment has 
 proved to be the actual product-volume resulting from the com- 
 bination of nitrogen and oxygen to form N^O, NO, and N0 2 re- 
 spectively. 
 
 Having thus shown that three out of the five terms of the 
 nitroxygen series obey the same law of condensation as that 
 observed in our typical series of bi-elementary compounds, it 
 remains to inquire whether the same normal condensation ratio 
 obtains in the two remaining terms, N 2 3 and N 2 5 . The 
 strongest analogies justify us in assuming that these bodies con- 
 form to the law ; but, owing to the special difficulties attending 
 their manipulation, the respective product-volumes of these two 
 compounds have not yet been determined by actual experiment. 
 
 It is therefore necessary, in the present state of science, to 
 distinguish the first, second, and fourth terms of the nitroxygen 
 series from the third and fifth, in respect to the evidence which 
 we possess as to their condensation ratios. This we may best 
 accomplish by drawing in full lines the squares employed to 
 express volumes determined by experiment, and using (as we 
 have done before) dotted lines to represent volumes at present 
 hypothetical, though assumed on the ground of analogies which 
 render their correctness in the highest degree probable. 
 
 This much premised, we may now graphically symbolize the 
 nitroxygen series, both as to the volumes of its constituents 
 and as to the product-volumes resulting from their condensation 
 in combining, as follows : 
 
OF THE NITROXYGEN SERIES. 119 
 
 VOLUMETRIC SYMBOLIZATION OF THE NITROXYGEN SERIES. 
 Nitrous Oxide. 
 
 N 
 
 N 
 
 Nitric Oxide. 
 
 Nitrous Acid. 
 
 O 
 
 Hyponitric Acid. 
 
 
 
 Nitric Acid. 
 
 
 
 To 
 
 
 
 
120 SUMMARY OF RESULTS. 
 
 On comparing the condensation ratios shown in this table 
 with those exhibited respectively in the symbolic diagrams of 
 our typical compounds (comp. pp. 70 and 106), we see that the 
 first and fourth terms of the series, N 2 and N0 2 , though con- 
 versely constituted, are counterparts in condensation, both ex- 
 emplifying the condensation ratio (f ) which we found to obtain 
 in the case of water (H 2 0). The second term (NO) exemplifies 
 combination without condensation (}-), as typified by hydrochloric 
 acid (HC1). The third and fifth, N 2 3 and N 2 5 respectively, 
 exemplify the more complex condensation ratios, f and ^ , for 
 which our previous experience affords us no precedents, but 
 whose complexity, I may mention in passing, we shall find very 
 much exceeded in numerous cases which will hereafter come 
 under our notice. These, however, are the two terms symbolized 
 by dotted squares, in token of their still hypothetical character ; 
 and great, indeed, would be my pleasure if any of those assem- 
 bled here should enable me to do away with these dotted lines, 
 and make these symbols full ones like the rest. 
 
 With these reserves as to the third and fifth terms of the 
 series, we may summarise as follows our present experience 
 of the 
 
 VOLUMETRIC COMPOSITION AND CONDENSATION OF BI-ELEMENTARY 
 CHEMICAL COMPOUNDS. 
 
 Constituents. Product. Condensation. 
 
 1 vol. 4- 1 vol. = 2 vols. 
 
 1 vol. 4 2 vols. = 2 vols. f 
 
 1 vol. 4- 3 vols. = 2 vols. 
 
 2 vols. + 3 vols. = 2 vols. f 
 2 vols. 4- 5 vols. = 2 vols. 
 
121 
 
 LECTURE VIII. 
 
 Transition from abstract to concrete formulae choice of a system of weights 
 and measures to supply units for the expression of concrete values 
 obstacles thereto, as also to the diffusion of science generally, by the want 
 of a universally-accepted sysfem of weights and measures French me- 
 trical system reasons for adopting it its general characters principles 
 of its nomenclature comparison with English measures the hydrogen 
 litre- weight, or criih the volume-weights of elements or compounds = the 
 absolute weights of 1 litre at 0' C. and O m '7G pressure, expressed in critlis. 
 
 THUS far, in our study of the weight and volume ratios trace- 
 able in the constitution of chemical compounds, we have referred 
 all our figures to hydrogen as unity, without complicating our 
 investigation by selecting some definitely-measured portion of 
 space to serve as a standard volume, and assigning to each ele- 
 mentary and compound gas its actual density, as determined by 
 weighing that volume in the balance. 
 
 To this inquiry, which will give to our formulae, hitherto 
 somewhat abstract, a more concrete, and therefore more practi- 
 cally useful, import, it is fitting that we should now proceed. 
 And to this end it is in the first place incumbent on us to 
 make choice of some one, among the many systems of weight 
 and measurement in use, to supply us with the needful standards 
 of gravity and dimension for reference. 
 
 And here I am fain to quit for a few moments my main topic, 
 and open a parenthesis, which shall be as brief as possible, for 
 the purpose of impressing upon your minds the obstacles pre- 
 sented to the cultivation and diffusion of scientific knowledge, 
 by the want of one universally accepted system of weights and 
 measures throughout the civilized world ; and by the unfortunate 
 prevalence, in lieu thereof, of a countless and confused variety 
 of special weights and measures, differing not only for every na- 
 tion and for every branch of human pursuits, but even for each 
 nation in its different provinces, and often in its petty towns and 
 
 G 
 
122 EVILS OF THE EXISTING CONFUSION OF WEIGHTS & MEASURES. 
 
 villages ; so that not only are the standards of measurement 
 multiplied, so as to fill bulky volumes with their mere enumera- 
 tion, but that standards, nominally the same, such as, for 
 example, the pound, the foot, the bushel, &c., have each of them 
 in different localities different values, counting by hundreds, 
 and probably, if all could be collected, by thousands. 
 
 It would be difficult to exaggerate the obstruction thus inter- 
 posed in the way of those who desire, by extensive inductions, 
 to collect and compare the experience of several nations, or the 
 statistics of several localities, so as to enlarge the basis of facts 
 on which all true science is founded. 
 
 Even to collate and compare the results of research in only 
 two or three countries, such as England, France, and Germany, 
 using different weights and measures, involves a perpetual labour 
 of reduction, and a gratuitous absorption of time and force, 
 deplorable when it is incurred, and still more deplorable when 
 it operates, as it too often does, to make the records of one 
 nation practically a sealed book to the philosophers of another. 
 
 I must not, however, dilate on these evils the mere cursory 
 indication of which is a digression from our immediate inquiry. 
 Enough has been said, if the wish has been inspired to terminate 
 as soon as may be the existing confusion, by promoting the 
 universal adoption of a Unitary System of weights and measures. 
 
 This leads us to inquire Which, among the numberless systems 
 in use, best deserves universal adoption ? or, in other words, 
 and to return to our more immediate subject. From which system 
 should we select our standards of volume and density, for the 
 purpose of assigning their concrete value to our hitherto abstract 
 symbols and formulae ? 
 
 To this question, on which, it must be admitted, considerable 
 differences of opinion prevail, my own answer is unhesitatingly 
 given in favour of the French metrical system ; the completeness 
 and simplicity of which has already won it extensive recognition 
 among the scientific communities of Europe, and incipient adop- 
 tion by several enlightened nations as their legal standard for 
 the ordinary purposes of life. It is, I believe, towards the 
 
FRENCH METRICAL SYSTEM ITS FUNDAMENTAL UNIT. 123 
 
 French metrical system that the balance of philosophic opinion 
 strongly inclines throughout Europe ; and this marked prepon- 
 derance is in itself a cogent argument, where unity of volition 
 and universality of practice are the very objects to be achieved. 
 I propose, therefore, in this place, to lay before you a concise 
 sketch of the French metrical system, which, as it seems des- 
 tined to come into general use throughout the world, so it has 
 ever been, and will continue to be, our standard of reference 
 throughout the whole series of chemical inquiries which we 
 have undertaken together. 
 
 The French metrical system is developed in all its ramifica- 
 tions, whether of linear, superficial, or solid dimensions, whether 
 of weight or capacity, from one primary element that funda- 
 mental unit being a measure of length ; and that standard 
 measure being sought in the simplest and sublimest of human 
 sciences, viz., in astronomy. This lineal unit is the 40-mil- 
 lionth part of the compass of our planet, as measured by a 
 circumference passing around it in a plane including its axis. 
 The circumference of the earth, thus taken, is termed a 
 meridian, or " great circle," and its 40-millionth part has 
 received the simple and well-chosen name metre (from the Greek 
 ^utYpo*', measure). From this single term, as from a common 
 root, all the ramifications of the French system well, therefore, 
 termed metrical are derived. Prolonged by decimal multipli- 
 cation, shortened by decimal subdivision ; in other words, 
 multiplied or divided by 10, by 100, by 1000, and so forth, the 
 metre supplies all the degrees of linear measurement, from those 
 required in the finest microscopic research to those employed to 
 span the firmament and denote the mighty courses of the stars. 
 Must we not, at the very outset, admit a certain grandeur in the 
 conception of a system, whose encyclopaedic scale supplies com- 
 mon and comparable expression for the dimensions of the 
 minutest atom, and those of interstellar space ? 
 
 As the lineal metre, decimally prolonged or shortened, fur- 
 nishes the universal linear scale, so the metre squared, and that 
 
 G 2 
 
124 GENERAL PRINCIPLES OF THE FRENCH METRICAL SYSTEM. 
 
 square decimally multiplied or divided, supplies all the grada- 
 tions of superficial measurement ; from the minute spaces of the 
 philosopher's micrometer, to those larger ones marked on the 
 land-surveyor's map, and to those still vaster areas in which the 
 geographer estimates the expanse of continents. 
 
 So, again, the square metre, simply cubed, and that cube 
 decimally multiplied and divided, supplies every gradation, 
 from minutest to mightiest, of the cognate measures of solidity 
 and capacity ; in other words, of space, considered as full or empty. 
 Multiply the cubic metre by one million, and you have a fit 
 measure in terms of which to express the capacity of the Atlantic, 
 or its cubical contents of brine ; divide the cubic metre by one 
 million, and you arrive at the petty volume of the gambler's ' 
 ordinary die. 
 
 This last-named volume, the millionth of the cubic metre, 
 taken as so much distilled water, furnishes the metrical unit 
 of weight, called the gramme a transition how admirable in its 
 simplicity! how useful, also, in forming, as it does, a link 
 between the volumetric and ponderal appreciations of material 
 quantity, in supplying for these different values cognate nume- 
 rical expressions, in bringing them thus within the reach of 
 easy, integral comparison (as contradistinguished from com- 
 parison by laborious fractional computations), and in facili- 
 tating, by these various means, the resolution of all the 
 problems, theoretical and practical, which nature presents for 
 solution by man ! 
 
 The gramme, decimally multiplied and divided, forms, in this 
 grand and simple system, the universal scale of weight. Its 
 millionth does not turn the finest balance ; its millionfold is the 
 unit for heavy merchandise (like the English ton) ; with its 
 thousandth commonly works the chemist ; with its thousandfold 
 the retail trader. The astronomer, when his business is to 
 weigh this or that fixed star, has only decimally to multiply the 
 raillionfold of the little gramme in order to derive from it a 
 weight-unit fitted for his use. Thus, by the adoption of this 
 unitary scale, are the majestic oscillations of the celestial orbs 
 
PRINCIPLES OF THE FRENCH METRICAL NOMENCLATURE. 125 
 
 rendered directly comparable with the vibrations of the che- 
 mist's granule in his balance. 
 
 From this general sketch of the French metrical system, we 
 must pass to take cognizance of its details, and particularly 
 to study the principles of its nomenclature. 
 
 These are as simple, and in their way as admirable, as the 
 system itself. It is only necessary to bear in mind the names 
 of the various units of Linear, Square, Cubic, and Ponderal quan- 
 tity ; as also to recollect that decimal prefixes taken from the 
 Greek are employed to signify multiplication, while decimal 
 prefixes taken from the Latin are used to express subdivision ; and 
 then, with these simple elements, the whole system, in all its 
 details, can be built up and mastered in a few minutes. 
 
 The Greek prefixes for 10, 100, and 1000, are respectively 
 deca, hecto, kilo. 
 
 The Latin prefixes for 10, 100, and 1000 are respectively 
 deci, centi, milli. 
 
 To the word metre, the unit of Linear dimension, let us apply, 
 first, the Greek prefixes which imply multiplication, secondly, the 
 Latin prefixes which imply subdivision. 
 
 By the first process we get the following series : 
 
 LINEAR MEASURE. 
 Unit 1 Linear Metre. 
 1. 'M.etie-multiples 
 
 Metres. 
 
 Metre = 1 
 
 Decametre = 10 
 
 Hectometre = 100 
 
 Kilometre = 1000 
 
 By the second process we obtain the series : 
 
 Metre. 
 
 Metre = I 
 
 Decimetre =0-1 or T ^ th of a metre. 
 
 Centimetre = O'Ol or 
 
 Millimetre = 001 or , 
 
126 FRENCH LINEAR AND SURFACE MEASURES. 
 
 We have thus, in the Greek series, suitable names for tenfold, 
 a hundredfold, and a thousandfold the unit ; while the Latin 
 series gives us appropriate appellations for its tenth, its 
 hundreth, and its thousandth part. 
 
 Of the Greek series, the first and last terms (the metre and 
 kilometre) are chiefly in use ; the former for such purposes as 
 the English yard subserves, the latter as road measure, in lieu 
 of the English mile. The intermediate terms are but little 
 required. 
 
 Of the Latin, or subdivisional series, on the contrary, all the 
 terms are in constant use, replacing respectively the English 
 yard, foot, inch, and various subdivisions of an inch (such as 
 its 8th, 10th, 12th, 16th, &c.). 
 
 Each of the other sorts of measures, to wit, Surface measure, 
 large and small, Cubic measure, Capacity measure, and Weight 
 measure, is shaped out on precisely the same model ; a conve- 
 nient unit of quantity being, in each case, selected to start from. 
 
 Thus, for Surface measure, on the large scale (in use for 
 measuring land), the metre square is deemed too small a unit 
 to set out with, so its first decimal multiple, the decametre, is 
 the linear unit selected for squaring, to obtain the primary 
 surface unit ; which is accordingly one square decametre, that 
 is to say, a square of 10 metres in the side, or 10 x 10 =100 
 square metres. This is called an area, or shortly an are. 
 Applying to this unit, as before to the metre, first the Greek or 
 multiplying, and then the Latin or subdividing prefixes, we 
 obtain the following tables of metrical 
 
 SURFACE MEASURE (LARGE SCALE). 
 
 Unit I Are. 
 1. Axe-multiples 
 
 Ares. Square metres. 
 
 Are 1 = 100 
 
 Decare = 10 = 1000 
 
 Hectare = 100 = 10000 
 
 Kilare = 1000 = 100000 
 
FRENCH WEIGHT-MEASURE. 127 
 
 2. Axe-divisions Are. Square metres. 
 
 Are = I = 100 
 
 Declare =0-1 10 
 
 Centiare = O'Ol 1 
 
 Milliare = 0-001 = 0-1 
 
 Of these measures the are and the hectare have come chiefly 
 into use. 
 
 The centiare of this series, it will be observed, corresponds 
 with the metre square ; and this, with its decimal subdivisions, 
 are employed for the measurement of small surfaces. Thus : 
 
 SURFACE MEASURE (SMALL SCALE). 
 
 Unit I Square Metre. 
 
 Square metre-divisions Metre square. 
 
 Metre square = 1 
 
 Decimetre square = O'Ol 
 Centimetre square = O'OOOl 
 Millimetre square = O'OOOOOl 
 
 As the unit of weight, the choice of the French has (as already 
 mentioned) judiciously fallen upon that die-sized cube already 
 alluded to as resulting from the subdivision of the metre cube of 
 distilled water (taken at 4C., its point of greatest density) into 
 a million equal parts. This they have called the gramme, from 
 ypa/x/xa, the name of a small weight in use among the Greeks. 
 The word ypa/x//,a is a derivative of ypa<o>, I write ; and it pro- 
 bably came into use for this purpose, from the circumstance that 
 the weights employed had their names, &c., inscribed upon 
 them. Tabulating, as before, with the aid of the Greek and 
 Latin prefixes, we obtain these two series : 
 
 WEIGHT MEASURE. 
 Unit 1 Gramme. 
 
 1. Gramme-mw^pZes Grammes. 
 
 Gramme 1 
 
 Decagramme = 10 
 Hectogramme 100 
 Kilogramme = 1000 
 
128 FRENCH CAPACITY-MEASURE. 
 
 2. Gr&mme-divisions 
 
 Gramme. 
 
 Gramme = 1 
 
 Decigramme = 1 or yV^ f a gramme 
 
 Centigramme = O'Ol or 
 
 Milligramme = 001 or 
 
 All the terms of this latter, or divisional series are found useful ; 
 but only the kilogramme, in addition to the unit itself, is found 
 requisite for ordinary use in the multiple series. The kilo- 
 gramme, or thousand -gramme-weight, is the weight of the cube 
 which results from dividing the metre cube into a thousand 
 equal parts : in other words, from dividing the edge of the 
 metre cube (. e. the metre) into 10 parts (i. e. 10 decimetres) and 
 cubing one of these parts (so as to obtain the decimetre cube). 
 This same decimetre cube is also adopted as the unit for mea- 
 sures of capacity, in which function it loses the name kilo- 
 gramme, and receives instead the appellation litre (from the term 
 \irpa, in use among the Greeks for the designation of one of 
 their standards of quantity). From this unit of capacity, as the 
 starting-point, the following capacity-measures are obtained by 
 decimal multiplication and division, distinguished by Greek and 
 Latin prefixes, as before : 
 
 CAPACITY MEASURE. 
 
 1. "Litre-multiples 
 
 Litres. 
 
 Litre 1 
 
 Decalitre = 10 
 
 Hectolitre = 100 
 
 Kilolitre = 1000 
 
 2. "Litre-divisions 
 
 Litre. 
 
 Litre = 1 
 
 Decilitre =0*1 or t^th of a litre. 
 
 Centilitre = 0*01 or y^-th 
 
 Millilitre = 001 or 
 
 All the measures in this table are employed alike for wet and 
 dry bodies (wine, grain, &c.), and they are all more or less in 
 
FRENCH CUBIC MEASURE. 129 
 
 use. The highest term (the kilolitre) corresponds to the metre 
 cube, the unit of solid measure. The lowest degree of the table, 
 or millilitre, is identical with the centimetre cube (*'. e. the 
 millionth of the metre cube, the die-like mass which, in distilled 
 water, represents one gramme), and it is under this name that 
 it is most commonly used. 
 
 It only remains to set forth the metrical cubic measures, 
 which start, like the linear measures, from the metre cube, and 
 form the two following series : 
 
 CUBIC MEASURE. 
 Unit 1 Cubic Metre. 
 
 1. Cubic metre-multiples 
 
 Cubic metres. 
 
 Cubic metre = 1 
 
 Cubic decametre = 1,000 
 
 Cubic hectometre = 1,000,000 
 
 Cubic kilometre = 1,000,000,000 
 
 2. Cubic metre-divisions 
 
 Cubic metre. 
 
 Cubic metre = 1 
 
 Cubic decimetre = O'OOl 
 
 Cubic centimetre = 0-000,001 
 
 Cubic millimetre = 0-000,000,001 
 
 The cubic metre is in common use as a measure of wood used 
 for fuel ; in which capacity it takes the appellation stere, from 
 the Greek arepfoc, solid. From this, as from the other units of 
 measure, a multiple and divisional series may be derived by aid 
 of the Greek and Latin prefixes ; but of the series so formed 
 only one member, the decistere, or tenth of a stere, has proved 
 needful, and come into use. 
 
 The highest term in each of the multiple series we have 
 tabulated is that with the prefix kilo. This may, however, in 
 each case, be decupled by substituting for kilo the prefix myria 
 (also Greek). Thus a myriametre is equal to ten kilometres ; 
 a myrialitre to ten kilolitres, and so forth. These high expres- 
 
 G 3 
 
130 COMPARISON OF FRENCH AND ENGLISH METRICAL UNITS. 
 
 sions, however, are comparatively seldom required ; and we have 
 kept our tables simpler by reserving this, their final term, for 
 separate mention, once for all, here at the close of our description. 
 It only remains to bring the units of this philosophical metric 
 system into comparison with those of the English arbitrary 
 scheme (the expression system would here be inappropriate) in 
 order simultaneously to display, first, the marked superiority of 
 the French over the English system ; secondly, the arithmetical 
 data for transforming the denominations of each into the other ; 
 and thirdly, the unwelcome amount of gratuitous toil attending 
 such elaborate reductions. In the following table the French 
 metrical units are placed in regard with, and reduced into terms 
 of, their respective English correlatives : 
 
 Metrical uniis. Yard lineal. Feet. Inches. 
 
 1 metre = 1-093633 = 3-280899 = 39-37079 
 
 Perches. Square yards. Acre . 
 
 lare = 3-957388 = 119-603326= 0' 02471 
 
 Grains. Pound. 
 
 1 gramme = 15-434000 - 0-002204 
 1 litre 
 
 1 stere 
 
 For our present purposes attention should be particularly 
 directed, first, to the units of weight and volume, namely, the 
 gramme and the litre, in terms of which the weight and volume 
 ratios with which we have to deal are best expressed ; and, 
 secondly, on the unit of length, the metre, in terms of which the 
 barometric pressure is best expressed, the average height of the 
 mercurial column being 0-76 metre 76 centimetres = 29-9218 
 English inches. 
 
 Closing here our parenthetical account of the metrical system, 
 and reverting to our immediate subject, we have now to select 
 an appropriate volume, with its corresponding weight, to serve as 
 our standard units of measurement. 
 
 = 
 
 Imp. pint. 
 
 1-760773 = 
 
 Yard cube. 
 
 1-308020 = 
 
 Imp. gallon. 
 
 0-220096 
 
 Cubic feet, 
 
 35-316580 
 
 Cubic inches. 
 
 = 61-027051 
 
Lineal Centimetres. 
 
 1 cubic decimetre, 
 
 = 1 litre, 
 
 = 1,000 cubic centimetres, 
 
 = T-^th of a cubic metre ; 
 
 containing, at 4C. temp 6 ', 
 
 grammes. 
 
 water, 1 kilo. = 1,000 ; 
 and, at 0C., & O m *76 press 6 - 
 
 gramme. 
 
 hydrogen, 1 crith = 0*0896 
 
 centi- 
 metre 
 
 To face page 131. 
 
CHEMICAL STANDARD VOLUME- WEIGHT-UNIT. 131 
 
 After much consideration, I am disposed to select 1 cubic 
 decimetre = 1 litre, as the most appropriate unit of volume for 
 our purpose ; and the weight of this measure of pure hydrogen 
 (the body in all respects best fitted to serve as a standard) as our 
 most suitable unit of weight. For this purpose hydrogen is 
 taken at 0C. temperature, and O m '76 pressure. 
 
 This volume-unit is represented by the cube shown on the 
 opposite page ; the front side of the cube being drawn of the 
 true size, the other sides receding in perspective. 
 
 The actual weight of this cube of hydrogen, at the standard 
 temperature and pressure mentioned, is O0896 gramme ; a figure 
 which I earnestly beg you to inscribe, as with a sharp graving 
 tool, upon your memory. There is probably no figure, in 
 chemical science, more important than this one to be borne 
 in mind, and to be kept ever in readiness for use in calculation 
 at a moment's notice. For this litre-weight of hydrogen = 
 0-0896 gramme (I purposely repeat it) is the standard multiple, 
 or co-efficient, by means of which the weight of 1 litre of any 
 other gas, simple or compound, is computed. Again, there- 
 fore, I say Do not let slip this figure : 0'0896 gramme. So 
 important, indeed, is this standard weight-unit, that some name 
 the simpler and briefer the better is needed to denote it. For 
 this purpose I venture to suggest the term with, derived from 
 the Greek word KpiOt), signifying a barley-corn, and figuratively 
 employed to imply a small weight. The weight of 1 litre 
 of hydrogen being called 1 crith, the volume-weight of other 
 gases, referred to hydrogen as a standard, may be expressed in 
 terms of this unit. 
 
 For example, the relative volume-weight of chlorine being 35'5, 
 that of oxygen 16, that of nitrogen 14, the actual weights of 1 litre 
 of each of these elementary gases, at 0C. and O m '76 pressure, 
 may be called respectively 35-5.cn'$s, 16 criths, and 14 critJis. 
 
 So, again, with reference to the compound gases, the rela- 
 tive volume-weight of each is equal to half the weight of its 
 product- volume. Hydrochloric acid (HC1), for example, con- 
 sists of 1 vol. of hydrogen -f 1 vol. of chlorine = 2 volumes ; 
 
132 THE HYDROGEN-LITRE-WEIGHT, OB CRITH. 
 
 or, by weight, 1 -f 35*5 = 3 6' 5 units ; whence it follows 
 that the relative volume-weight of hydrochloric acid gas is 
 
 OC .K 
 
 -- = 18-25 units; which last figure, therefore, expresses the num- 
 ber of criths which 1 litre of hydrochloric acid gas weighs at 0C. 
 temperature and O m -76 pressure ; and the crith being (as I 
 trust you already bear in mind) 0-0896 gramme, we have 
 18-25 X 0-0896 = 1-6352, as the actual weight in grammes 
 of hydrochloric acid gas. 
 
 So, once more, as the product-volume of water-gas (H 2 0) 
 (taken at the above temperature and pressure) contains 2 
 vols. of hydrogen + 1 vol. of oxygen, and therefore weighs 
 
 2 4- 16 = 18 units, the single volume of water-gas weighs 
 
 18 
 
 = 9 units ; or, substituting as before the concrete for the ab- 
 
 2> 
 
 stract value, 1 litre of water-gas weighs 9 criths ; that is to say, 
 9 x 0-0896 gramme = 0-8064 gramme. 
 
 In like manner the product-volume of sulphuretted hydrogen 
 (H 2 S) = 2 litres of hydrogen, weighing 2 criths + 1 litre of 
 sulphur-gas, weighing 32 criths, together 2 + 32 = 34 criths, 
 
 4 
 which divided by 2 gives -jr- = 17 criths = 17 x 0*0896 gramme 
 
 2 
 
 = 1-5232 gramme = the weight of 1 litre of sulphuretted hy- 
 drogen at standard temperature and pressure. 
 
 And so, lastly, of ammonia (H 3 N) : it contains in 2 litres 3 litres 
 of hydrogen, weighing 3 criths, and 1 litre of nitrogen, weigh- 
 ing 14 criths; its total product-volume-weight is therefore 
 
 3 + 14 = 17 criths, and its single volume, or litre weight is 
 
 17 
 consequently = 8-5 criths = 8-5 x 0-0896 gramme = 07616 
 
 A 
 
 gramme. 
 
 Thus, by aid of the hydrogen-litre-weight or crith =0-0896 
 gramme, employed as a common multiple, the actual or concrete 
 weight of 1 litre of any gas, simple or compound, at standard 
 temperature and pressure, may be deduced from the mere 
 
VALUE OF THIS STANDAKD. 133 
 
 abstract figure expressing its volume- weight relatively to hy- 
 drogen. 
 
 From this knowledge, the weight of 1 litre of any gas, simple 
 or compound, at any other than standard temperature or pres- 
 sure, or under any variation both of standard temperature and 
 pressure, may be deduced, by the application of the formulas 
 devised by physicists to express the laws ,of expansion and con- 
 traction, for gases under varying conditions of temperature and 
 pressure. The volume is, for gases, inversely as the pressure ; 
 and every degree of temperature, added or subtracted, implies a 
 certain fractional expansion or contraction of a gas's bulk. 
 Diminished pressure and increased temperature consequently 
 lighten the litre of any gas, in perfectly well-known and definite 
 proportions. Hence, from the weight of 1 litre of gas, taken at 
 any temperature, however high or low, the weight of 1 litre 
 taken as at 0C. can be unerringly deduced by computation. 
 
 This fact is of great value, as enabling us to include in com- 
 mon and comparable forms of expression the gas-volume-weights 
 of bodies that are not, as well as of those that are, gaseous at 
 O 1 centigrade. Bodies which are liquid, or even solid, under 
 this standard physical condition, have only to be weighed at the 
 temperature, however elevated, at which they do assume the 
 gaseous form, and from the weight of 1 litre, under this physical 
 condition, the weight which 1 litre would assume at the standard 
 temperature, could it remain gaseous thereunder, is deducible 
 by computation. 
 
 The correctness of the weight and volume ratios thus ob- 
 tained can be put to the test, conversely, by subjecting hydro- 
 gen to the same temperature as that which is needful to raise 
 into gas the less volatile body with which we may desire to 
 compare it. The physical conditions of attenuation being 
 thus rendered equal for both bodies, their respective volume- 
 weights may be experimentally determined, and the ratio of 
 these recorded. In this way hydrogen, bromine, and iodine 
 have been compared as to their relative volume- weights, at the 
 temperatures needful to convert the two latter bodies into gas ; 
 and the values thus obtained have been found to stand to each 
 
134 CONCRETE GAS-VOLUME-WEIGHTS, 
 
 other in the ratios 1 : 80 : 127 for hydrogen, bromine, and 
 iodine respectively. In other words, to whatever volume the 
 litre, or crith, of hydrogen may be dilated, that same volume of 
 bromine and iodine, subject to like conditions of dilatation, will 
 weigh 80 criths and 127 criths respectively. 
 
 Again, bodies which, like carbon, cannot be volatilized alone 
 under any physical conditions whatsoever, but which become 
 gaseous on combining with certain aeriform bodies as, for in- 
 stance, in the case of carbon, on combining with hydrogen may 
 be brought hypothetically under our general volumetric and 
 ponderal expressions, by computations, based on analogy, and 
 affording, not indeed certain, but more or less probable re- 
 sults. Thus 2 litres of marsh-gas are found to contain four 
 litres of hydrogen, which, according to analogy, may be pro- 
 bably combined either with 1 litre of carbon gas, or, as in 
 the exceptional cases of phosphorus and arsenic, with half that 
 volume. As the weight of the carbon combined with 4 litres = 
 4 criths of hydrogen is known by analysis to be 12 criths, the 
 litre of carbon gas, at standard temperature and pressure, pro- 
 bably weighs either 12 or 24 criths, accordingly as its deport- 
 ment is assumed to resemble that of the majority of volatile and 
 volatilisable bodies, or to coincide with the exceptional beha- 
 viour of phosphorus and arsenic. The former assumption has 
 been adopted by many chemists for the purpose of provisionally 
 assigning to carbon a place in the general system of volumetric 
 and ponderal ratios ; but this view, it must be borne in mind, is 
 entirely speculative. 
 
 With these additions to our stock of knowledge, the symbols 
 of the volatile elements, and the formulae of the volatile com- 
 pounds, acquire for us a new significance. We have already seen 
 that they express the single volume-weights of all the elementary 
 bodies symbolized, with the exception of phosphorus and 
 arsenic, of which they express the half volume-weights only. 
 We have also seen that, of the compound bodies, they express, 
 in all cases, without exception, the product-volume-weights, so far 
 as these have as yet come under our notice. We now find that, 
 
EXPRESSED IN TERMS OF THE CRITH, AND OF THE GRAMME. 135 
 
 with the aid of one co-efficient, the key to them all, these abstract 
 weights may be transformed into the corresponding concrete or 
 actual weights at 0C. temperature, and O m -76 pressure. That key, 
 or co-efficient, is the crith = 0-0896 gramme, and I think you 
 will now see the grounds of my often-repeated hope that you 
 will firmly commit it to memory. 
 
 TABLE OF GAS- VOLUME- WEIGHTS. 
 
 1 
 
 2 
 
 3 ' 
 
 4 
 
 
 
 Gas- vol. - 
 
 
 
 
 weights, or 
 
 
 NAMES. 
 
 Symbols 
 and 
 Formulae. 
 
 weights of 
 one litre 
 at 0C. and 
 O m - 76 Bar. 
 = sp. gr. 
 relatively 
 to hydro- 
 
 Gas- vol. -weights, 
 or weights of one 
 litre at 0C. and 
 O m - 76 Bar. = sp. 
 gr. relatively to 
 hydrogen in terms 
 
 
 
 gen in 
 
 of a gramme. 
 
 
 
 terms of a 
 
 
 
 
 crith. 
 
 
 SIMPLE BODIES. 
 
 
 
 
 Hydrogen 
 
 H 
 
 1 
 
 0-0896 
 
 Chlorine 
 
 Cl 
 
 35-5 
 
 3-1808 
 
 Oxygen 
 
 O 
 
 16 
 
 1-4336 
 
 Nitrogen 
 
 N 
 
 14 
 
 1 2544 
 
 COMPOUND BODIES. 
 
 
 
 
 Typical. 
 
 
 
 
 Hydrochloric acid .. .. 
 
 HC1 
 
 18-25 
 
 1-6352 
 
 Water-gas (dry steam) .. 
 
 H 2 
 
 9 
 
 0-8064 
 
 Ammonia 
 
 H 3 N 
 
 8-5 
 
 0-7616 
 
 Marsh-gas 
 
 H 4 C 
 
 8 
 
 0-7168 
 
 Nitroxygen Series. 
 
 
 
 
 Nitrous oxide 
 
 N 2 
 
 22 
 
 1-9712 
 
 Nitric oxide 
 
 NO 
 
 15 
 
 1 3440 
 
 Nitrous acid 
 
 NA 
 
 38 ? 
 
 3-4048 
 
 Hyponitric acid 
 
 NO 2 
 
 23 
 
 2-0608 
 
 Nitric acid 
 
 NA 
 
 54 ? 
 
 4-8384 
 
 In the above table, the first column contains the names of 
 
136 VALUE OF THE CO-EFFICIENT 0'0896 GRAMME. 
 
 the simple bodies and their compounds, so far as our studies 
 have yet extended ; in the second column are placed their sym- 
 bols and formulae ; in the third, their gas- volume weights, or 
 specific gravities relatively to hydrogen at standard temperature 
 and pressure, read in terms of one crifh ; while the fourth and 
 last column gives the same values expressed in terms of one 
 gramme. 
 
 For many purposes, where closely-approximative results suf- 
 fice, trouble in multiplying may be saved, by employing O090 
 gramme instead of 0*0896 gramme as the value of the crifh. The 
 error for chlorine (the heaviest of the elementary gases) is only 
 3-1950 - 3-1808 = 0-0142 gramme in excess on the weight of 
 the litre ; for hydrogen (the lightest body), it is only 4-ten 
 thousandths of a gramme. 
 
 I will only ask your attention to-day to one further remark on 
 this subject, having reference to a point of difference worth 
 noting between the symbols of the elementary and the formulae 
 of the compound gases. Eead in criths, the former expressions 
 give the weights of 1 litre, the latter of 2 litres, of the ele- 
 mentary and compound gases respectively ; so that we may pro- 
 perly term the former monolitral, and the latter dil'dral^ expres- 
 sions. In the exceptional cases of phosphorus and arsenic, the 
 symbols, read in criths, give the weight of half volumes only of 
 their respective vapours ; so that these are Ticmilitral expressions. 
 These discrepancies are attended with some inconvenience, 
 which, however, I may mention in passing, the course of our 
 future inquiries will enable us, in a great measure, to obviate. 
 
 The unit crith might, of course, be multiplied and divided by 
 the Greek and Latin prefixes like any of the units of the metri- 
 cal system ; but no useful purpose would be served thereby. 
 Indeed, the term crith may be dispensed with altogether by those 
 whose memory can retain, without assistance, the value of the 
 co-efiicient 0-0896. 
 
137 
 
 LECTUKE IX. 
 
 Philosophical conceptions of chemical phenomena hypotheses theories 
 matter its nature and essence its conditions, solid, fluid, gaseous its 
 activities, molar and molecular properties of molecules their mutual 
 cohesion and repulsion nature and properties of gases their elasticity 
 their latent heat nature of heat molecular dynaspheres observed in- 
 fluence of temperature and pressure on gases identical comportment of 
 all gases under like variations thereof composite structure of molecules 
 conception of atoms how arrived at final term of the known threefold 
 divisibility of matter its infinite divisibility, why not to be affirmed 
 elementary molecules diatomic molecules tetratomic molecules their 
 symbolization atomic and molecular forms of notation molecular sym- 
 bolization exemplified in the formulae of the nitroxygen series compa- 
 rative advantages of the atomic and molecular form of notation. 
 
 AT the present stage of our inquiry, a new aspect of chemical 
 phenomena presents itself to our notice, and its due considera- 
 tion will justify us in assigning to the symbols of the elements, 
 to the formulae of the compounds, and to the equations express- 
 ing their reactions, a deeper signification than any they have 
 hitherto possessed for us. 
 
 Up to the present moment we have not quitted the domain of 
 experience ; we have confined our attention to facts, either wit- 
 nessed by ourselves, or accepted on the testimony of others. 
 We have been satisfied to observe chemical phenomena, with- 
 out seeking to explain them, save in so far as their orderly 
 collection, comparison, and record may be held to constitute ex- 
 planation. 
 
 To the causes of the remarkable effects we witnessed, our 
 attention has not yet been turned. Yet the inquiry into the 
 causes of observed phenomena is urged on us by one of the 
 strongest instincts of our intellectual nature. That instinctive 
 
138 PHILOSOPHICAL CONCEPTIONS HYPOTHESES THEORIES. 
 
 curiosity cannot, indeed, be fully satisfied. The first causes of 
 phenomena lie beyond the limited scope of our perceptive and 
 reasoning faculties. The conditions of their existence or produc- 
 tion, and their relations of succession and similitude are, indeed, 
 open to investigation; but their intimate nature and prime 
 origin are for us inscrutable mysteries. We may, however, by 
 the aid of imagination, form hypotheses (a word of Greek origin, 
 from {ITTO, under, and 0ecris, a derivative of riO^i, I place, cor- 
 responding to the Latin suppositio, from sub, under, and positio, 
 a derivative ofpono, I place, whence the English word supposition) 
 to connect the results of our experiments and to guide the 
 course of our inquiries. And, though merely speculative hypo- 
 theses, dissevered from experimental investigation, are to be 
 deprecated as vain and sterile exercises of ingenuity, hypotheses 
 based upon facts, assisting in their conception, and deriving pro- 
 bability from the number thereof which they connect and ex- 
 plain, besides (and above all) tending to suggest new experi- 
 ments, deserve to rank among the most valuable aids to scientific 
 research. 
 
 Hypotheses are, of course, to be held provisionally, subject 
 to modification and abandonment, in so far as they may from 
 time to time prove inconsistent with the results of further 
 experimental research. On the other hand, when hypotheses 
 embrace and explain extensive ranges of phenomena, when 
 experiment confirms the results they foreshadow, when successive 
 discoveries raise them higher and higher in the scale of proba- 
 bility, they lose more and more their provisional character, 
 and gradually assume the name and rank of theories (from the 
 Greek ^eoopew, / contemplate), till at last they come to be 
 embodied permanently among the recognized doctrines of philo- 
 sophy and science. 
 
 The observed phenomena of combination in definite propor- 
 tions by weight and volume are susceptible of explanation by a 
 theoiy in the highest degree probable and suggestive, which the 
 experiments we have made together, and the symbols in which 
 we have recorded their results, have prepared us to understand. 
 
ESSENCE AND NATURE OF MATTER ITS DIVISIBILITY. 139 
 
 To this theory we shall now devote some attention, without, 
 however, venturing too far as yet upon speculative ground. 
 
 In order to arrive at the theoretical conception in question, 
 we must ask ourselves, What is matter? Of what parts is it 
 composed? How are these constructed and held togeth'er? 
 How comes the very same matter, water for example, to pre- 
 sent itself sometimes in the solid form, as ice ; sometimes in the 
 liquid form, as the same ice when melted; sometimes in the 
 gaseous form, as the same melted ice changed to dry steam by 
 further heating? And, lastly, what happens to matter, what 
 changes does it undergo, when its various elementary forms 
 combine, as we have seen them, to produce bodies having pro- 
 perties wholly different from those of their constituents ? 
 
 These and other analogous questions have occupied philoso- 
 phers during many generations ; and the briefest history of the 
 innumerable controversies thus engendered would fill many 
 volumes. Even the single preliminary inquiry Is matter in- 
 finitely divisible, or does it consist of smallest particles, in- 
 capable of further subdivision ? has given rise to contradictory 
 opinions and arguments, which merely to summarize would 
 largely exceed the limits of time and space at our disposal. 
 
 Experience of the value of hypothetical conceptions, when 
 confined within just limits, and of their sterility when pushed 
 too far, has imbued the leaders of modern philosophy with a 
 spirit of extreme moderation and reserve in their attempts to 
 penetrate these deep secrets of nature. 
 
 Setting aside, in a similar spirit, the more transcendental 
 speculations of philosophers upon the nature of matter, let us 
 here select for consideration those hypothetical conceptions of 
 its structure which seem best adapted to connect and explain 
 the results of modern research ; and which, by enabling us to 
 comprehend the phenomena we have already witnessed, may 
 also assist us in shaping the course of our further experimental 
 researches. 
 
 Let us, for this purpose, consider the familiar body, water, 
 into the nature of which our experiments have already given us 
 
140 WATEK SOLID, FLUID, GASEOUS. 
 
 some insight ; and let us consider it in its three conditions, as 
 ice, as fluid water, and as water-gas or dry steam, all which are 
 here before us. What is the first thing that strikes us in look- 
 ing at them ? 
 
 The first thing that strikes us is, that ice, water, and steam 
 manifest two sorts of activity one exerted by masses of sensible 
 magnitude, acting through measurable distances of space ; the 
 other operating between particles, and through intervals of 
 space, so minute as to be incommensurable. 
 
 The attraction of mass for mass of matter, which we see 
 manifested in the courses of the celestial bodies, in the move- 
 ment of falling bodies, and in the pressure of bodies at rest 
 upon the ground, exemplifies the first kind of activity. This is 
 equally observable in the ice, in the water, and in the water- 
 gas; for these all possess weight; a sensible mass of either 
 reciprocates attraction with the earth, through measurable dis- 
 tances of space. 
 
 The Latin for mass is moles; and its modern diminutive, 
 molecula, is employed to designate "a little mass," that is to 
 say, a material particle of incommensurable minuteness ; hence 
 the reciprocal actions of minute particles through insensible 
 intervals of space are distinguished as molecular. We may 
 fairly therefore contradistinguish, by the epithet molar, the 
 reciprocal actions of measurable masses through measurable 
 intervals of space. 
 
 The means of mechanical comminution at our disposal, our 
 grinding-mills, mortars, and the like, do not carry us beyond the 
 molar subdivision of matter. However finely we might grind up 
 this ice, for example, if we took care to keep the temperature below 
 freezing-point, we should still have masses consisting of several 
 molecules. For, our finest ice-powder would still consist of 
 very small fragments of solid ice ; and if, of this ice-dust, we 
 took the smallest grain, we could, by applying heat, turn it into 
 water, thus proving it to have parts, capable of separation, so as 
 to be rendered moveable amongst each other. There is no 
 instance of liquefaction resulting from the mechanical comminu- 
 
DISTINCTION OF MOLAR FROM MOLECULAR ACTIVITIES. 141 
 
 tion of a solid body. Hence we take it as certain that the most 
 impalpable product of mechanical pulverization is still a cluster 
 of molecules. 
 
 We are thus enabled to distinguish in matter two kinds of 
 divisibility, molar and molecular ; the former being accomplished 
 by mechanical means, and only resulting, even when pushed to 
 its utmost attainable limits, in the production of a molecule- 
 cluster or mass of- sensible dimensions, which may be termed a 
 mole; while the latter is accomplished by physical means (that 
 is to say, by aid of physical forces, such as heat), resulting in 
 the disruption of the masses or moles into their incommensurably 
 minute constituent molecules. 
 
 The study of the reciprocal action of material masses, or 
 moles, constitutes the science of mechanics; a science of the 
 deepest interest, abounding in simple and admirable laws, with 
 which, however, we are not at present concerned. 
 
 Turning to the consideration of molecular activities, of those 
 which are distinguished by the incommensurable minuteness of 
 the particles of matter, and of the intervals of space, between and 
 through which they take place ; and looking once more at the 
 samples of matter before us at our ice, our water, and our 
 water-gas or steam ; we are again, as before, struck with a con- 
 trast between two diametrically opposite kinds of activity, one 
 conspicuously manifested in the solid ice, and called molecular 
 cohesion, the other especially manifested in the water-gas, and 
 termed molecular repulsion. The former force gives to solid 
 bodies their tenacity; to the latter, gaseous bodies owe their 
 extreme tenuity, and the free mobility of their molecules 
 amongst each other. 
 
 In fluid bodies, here represented by our water, we observe 
 these two forms of molecular activity balanced at an intermediate 
 point. The molecules of fluids cohere with considerable force ; 
 as you perceive, when I dip this rod into the water, and take 
 out a bunch of them, sticking to each other, and also to the rod, 
 in the form of this pendent water-drop ; but this cohesion is 
 exceedingly feeble as compared with that of the similar mole- 
 
142 MOLECULAR ATTRACTION AND REPULSION. 
 
 cules agglomerated in the solid form here in our block of ice. 
 Again, the molecules of fluids are moveable amongst each other ; 
 as you notice when I shake the water in this vessel, when I 
 agitate it with this rod, and when I pour it into this other glass ; 
 but their mobility is far inferior to that of the molecules of gas. 
 In vain should we dip our rod into the gas to take up a drop of 
 it ; we should obtain no coherent bunch of gas-molecules, like 
 our pendulous water-drop. And it is precisely to their superior 
 molecular cohesion that fluids owe their inferior molecular 
 mobility as compared with gases. Hence the property of fluids 
 termed their viscidity, a property which varies greatly in dif- 
 ferent fluids, so as to render unequal (for example) their rate of 
 flow through tubes; but from which all gases are absolutely 
 free, their molecules tending rather to recede from each other 
 than to cohere. The only property which, in this respect, at all 
 assimilates gaseous to fluid bodies, is the tendency of the former 
 to adhere in thin films to the surfaces of solid bodies dipped 
 into them or otherwise brought into contact with them. In this 
 way a gas may be said to wet a solid body just as a fluid does i 
 but here the resemblance ceases. For, in the case of the fluid, 
 the inter-molecular adhesion, or viscidity, causes the first film to 
 attract another film, and this a third, and so on, till a coating of 
 sensible thickness is obtained ; whereas, in the gas, the original 
 film attracted by the solid does not in its turn attract a second, 
 but remains of insensible thickness. How its presence is made 
 known we shall hereafter learn from interesting experiments. 
 At present these phenomena only concern us, as marking the 
 different behaviour of matter, accordingly as its condition is 
 fluid or gaseous. 
 
 This difference of comportment is not surprising when we 
 reflect how much greater are the intervals which separate the 
 molecules of a gas of our water-gas, for example than those 
 which intervene between the molecules of the same body in the 
 form of ice or of water. 
 
 Ice and water differ very little in bulk ; at and near the freez- 
 ing-point, indeed, water occupies rather less space, weight for 
 
STRUCTURE OF GASES THEIR ELASTICITY. 143 
 
 weight, than ice. But water-gas or steam, at 100 C., occupies 
 1689 tiroes more space than water at the same temperature. 
 
 Hence it follows that the molecules of water-gas are separated 
 by intervals which, though doubtless incommensurably small, 
 are nevertheless 1689 times larger than those which separate the 
 same molecules when reduced to the fluid condition. 
 
 For the student of volumetric chemistry the structure of gases 
 has a special interest ; and it is on this branch of the inquiry into 
 the nature of matter that we shall here bestow our principal 
 attention. 
 
 What is the nature of the intervals between the molecules of a 
 gas ? are they empty space, or are they filled ? and, if so, how, or 
 with what are they filled ? That they are not empty spaces we 
 have very good reason to believe, on account of the powerful 
 resilient property manifested by gases when forcibly compressed. 
 Here, for example, is a moist bladder full of a gas, of. common 
 air you observe how powerfully it resists my forcible en- 
 deavours to compress it ; and, when I strike it with this mallet, 
 you see, by the bounding back of the implement, the powerful 
 resilience, or elasticity, which gas possesses. 
 
 But what is the nature of this elasticity or resilience to 
 what power or force is it due ? 
 
 Several phenomena point to heat as its cause. Heat is the 
 agent by which ice is made to pass, through the fluid, into the 
 gaseous form; and, with every increment of heat, the elastic 
 power of the ice-derived gas augments. This -is true, not of 
 water-gas only, but of all gases ; and of this we can easily obtain 
 illustration by heating our bladder full of common air : you per- 
 ceive how much tenser it becomes as the gas absorbs heat : when 
 I strike it as before the mallet, you observe, is more powerfully 
 resisted ; and, upon heating it still more, its elasticity overcomes 
 the tenacity of the envelope, and it bursts with the loud report 
 you hear. 
 
 Again, if I set over a gas-burner a capsule containing a lump 
 of ice of the temperature of C., the ice gradually melts, and 
 becomes converted into water ; but if I try with a thermometer 
 
144 LATENT HEAT OF LIQUIDS OF GASES. 
 
 the temperature of the water at the moment when the last 
 particle of ice is melted, I find it still at C. That the capsule 
 has been absorbing heat we are sure; first, because of its 
 position over the flame ; secondly, because the ice it contains has 
 melted. 
 
 What has become of the heat thus absorbed ? and how comes 
 it that the water produced, when heat melts ice, is nevertheless 
 an ice-cold fluid ? 
 
 If any one be tempted to doubt the absorption of heat by the 
 capsule, on account of the fact that the water it contains is 
 ice-cold, nothing is easier than to convince him of his error. 
 For this purpose it suffices simply to leave the capsule, with its 
 ice-cold contents, in its place over the flame ; and the thermo- 
 meter will soon bear witness to a continuous and rapid absorp- 
 tion of heat. The water, ice-cold at starting, grows hotter and 
 hotter every minute. The rate of increase of the heat shows 
 the heat-absorbing power of the capsule ; so that, by noting how 
 many minutes the ice takes to melt, we know how much heat 
 was absorbed during the process. Meanwhile, the thermometer 
 plunged into the water continues to rise, till at last it reaches 
 100 C. ; and then, as we all know, the water boils. 
 
 If, now, we still leave the capsule in its place, it continues, of 
 course, to imbibe heat at the same rate as before ; but the water 
 now once again ceases to increase in temperature. It merely 
 " boils away," as we familiarly say ; that is, it becomes progres- 
 sively converted into water-gas, which escapes by diffusion into 
 the air, till at last all the water disappears. 
 
 If we now leave the empty capsule over the flame, we have 
 soon another proof of the heat it has been imbibing all this time ; 
 for its bottom speedily becomes red-hot. 
 
 Now, precisely as much heat as, in a given time, entered the 
 capsule to make it red-hot, precisely so much, in each equal 
 interval of time, entered the ice to turn it into water, and en- 
 tered the water to turn it into steam. Yet the water was not 
 hotter than the ice just melted, ner the steam than the water 
 just on the point of volatilization. Evidently, therefore, heat 
 
HYPOTHESIS OF MOLECULAR DYNASPHERES. 145 
 
 enough to make the bottom of the capsule red-hot several times 
 over has entered into the water and into the water-gas. Yet it 
 has not made these red-hot. What has become of the heat thus 
 absorbed and hidden ; or, to use the scientific expression derived 
 from the Latin, thus rendered latent ? We see, concurrently 
 with its disappearance, the molecules of ice loosen into fluid 
 water, and the molecules of water take wing and form resilient 
 gas. I could even give you, did time permit, experimental 
 proof that, by condensing the water-gas back into water, we 
 can recover the hidden force as sensible heat again ; and, in like 
 manner, that to freeze the condensed water so obtained into ice> 
 there must be withdrawn from it just so much heat as the ice, by 
 whose liquefaction we obtained it, absorbed and rendered latent in 
 melting. Evidently then, putting these things together, we 
 cannot help connecting the absorption and latency of the heat 
 with the successive development of the fluid and gaseous condi- 
 tions ; nor can we regard the elasticity of the steam otherwise 
 than as a result due to some form or modification of the force 
 known to us as heat. 
 
 To the questions, therefore, what is a gas ? and with what are the 
 intervals between its molecules filled? succeeds the question, 
 what is heat ? This brings us face to face with one of the most 
 ardently-mooted and deeply-interesting philosophical questions 
 of the day. For some, heat is a species of thin ether, vibrating 
 in the manner of light ; for others, it is a pure force, having 
 neither parts nor weight ; for a third class of thinkers, of late 
 years the majority, heat has no separate existence, but is merely 
 a mode of motion, the result of the vibration of material mole- 
 cules. 
 
 It is no part of our present task to attempt the solution of this 
 deep and difficult problem. We may content ourselves here with 
 the conception that heat, whatever may be its intimate nature, 
 so operates, when it becomes latent in a gas, as to surround 
 each molecule with a sort of repellent atmosphere which tends 
 to keep it apart from its fellows; and that these molecular 
 force-spheres or, to employ the Greek equivalent, dynami-spJieres, 
 
146 COMPOKTMENT OF SIMPLE AND COMPOUND GASES 
 
 more shortly, dyna-spheres (from the Greek SvW/xis, force, a 
 derivative of Suva/oat, I can) when mechanically compressed, 
 counteract the pressure with exactly equal energy, and on the 
 removal of the pressure, restore the gas (other things being 
 equal) to the exact volume it previously possessed. 
 
 Now, carrying on observation side by side with hypothesis, 
 let us compare with the behaviour of water-gas, under varied 
 conditions of temperature and pressure, that of the other com- 
 pound gases, hydrochloric acid gas, ammonia, and marsh-gas, 
 and also that of our four simple or elementary gases, hydro- 
 gen, chlorine, oxygen, and nitrogen ; or, to obviate the difficulty 
 which attends the study of water-gas, on account of its falling 
 back into the fluid state at ordinary temperatures, let us select, 
 from among the more permanent compound gases, some type to 
 represent them, say hydrochloric acid ; and in like manner let us 
 select from among the simple gases a typical representative, say 
 hydrogen ; and let us submit these two representative gases to 
 equal increments and decrements of temperature and pressure 
 successively. If they behave differently, we shall have reason 
 to believe that the molecular structure of elementary and com- 
 pound gases is different ; }f they behave alike, we shall be justi- 
 fied in assuming their molecular structure to be the same. 
 
 We have here an apparatus fitted up for the purpose of making 
 this comparison rapidly, and on a scale easily visible. It con- 
 sists of a modification of the double U tube, used already for a 
 different purpose (comp. p. 53) ; and it has one long and simple 
 limb, and one short limb bifurcated into two branches, each of 
 which is provided with a stopcock. These short limbs are, 
 moreover, enclosed in a cylinder of glass, as shown in the figure. 
 The purpose of this glass envelope will appear in the sequel. 
 Near the bottom of the apparatus another stopcock will be ob- 
 served, so placed as to serve for emptying it. The whole appa- 
 ratus being filled with mercury, the gases to be compared may be 
 easily drawn, each into one of the two small branch-limbs, by 
 connecting these, through flexible tubes attached to their 
 stopcocks, with the apparatus supplying the respective gases, 
 
UNDER LIKE VARIATIONS OF TEMPERATURE AND PRESSURE. 147 
 
 and then letting out the mercury through the stopcock 
 below. Into the vacuum which the descent of the mercury 
 
 FIG. 6-K 
 
 leaves in the branch-tubes, the gases to be studied of course 
 flow; and, when the branches are thus about half filled, the 
 stopcocks are closed ; care however being taken to have a 
 
 H 2 
 
148 IDENTICAL MOLECULAR STRUCTURE OF ALL GASES. 
 
 precisely equal quantity of gas in each of the two limbs. This 
 adjustment made, we may mark with caoutchouc rings the space 
 filled with gas in each tube. Upon pouring more mercury 
 into the long tube, we obtain a column which exercises precisely 
 equal pressure on the gases in the two branches ; and we see 
 that these are equally compressed, by the equal ascent of the 
 mercury in each branch. On the other hand, on letting out 
 mercury through the bottom stopcock, so as to lower the column 
 and decrease the pressure, we observe that both gases, different 
 as they are by nature, undergo equal dilatation. And so, again, 
 the expansion and contraction of the two gases is shown to be 
 equal, under equal increments and decrements of heat, by filling 
 the cylinder which surrounds the bifurcated limbs of the appa- 
 ratus with an atmosphere of hot steam or cold air by turns. 
 
 It thus stands clearly demonstrated that, if equal volumes of the 
 elementary gas, hydrogen, and of the compound gas, hydrochlo- 
 ric acid, be taken under any given temperature and pressure, 
 and the pressure be doubled for each, each becomes reduced 
 to half its former volume, and at the same time acquires double 
 its former resilient force, or elasticity ; which it exerts in coun- 
 terbalancing the pressure from without. 
 
 It stands equally proved that, if equal volumes of hydrogen, 
 and of hydrochloric acid gas, taken at equal degrees of pressure 
 and temperature, be exposed to equal increments or decrements 
 of heat, they undergo equal degrees of expansion and contrac- 
 tion. 
 
 Experiments of this kind, repeated for numerous gases, simple 
 and compound, have established as a law, that all true gases, 
 simple as well as compound, comport themselves in sensibly 
 the same manner under like variations of temperature and 
 pressure ; whence the inference fairly follows that their mole- 
 cular structure is the same. Assuming, then, each gaseous 
 molecule to be clothed or enveloped by a resilient dynasphere 
 (as we have termed it), due, in some unknown way, to the 
 influence of latent heat, experiment justifies us in inferring, 
 from the identical comportment of all gases, when exposed to 
 
3IIXUTEXESS OF THE MOLECULES OF GASES. 149 
 
 like variations of temperature and pressure, that they all contain, 
 in equal volumes, an equal number of molecules so clothed ; and 
 that, as an obvious corollary, the diameter of these gas-mole- 
 cules (including in that term as well the dynaspheres as their 
 material nuclei) is, under like physical conditions, precisely the 
 same for all gases. To express it more shortly, our unit-volume, 
 or litre, whether of hydrogen, of hydrochloric acid, or of any 
 other gas, simple or compound, is composed of mutually repel- 
 lent dynaspheric molecules, equal (omnibus paribus) as to their 
 number, and (consequently) as to their size. 
 
 At this point of our inquiry we may advantageously resume 
 the consideration of material divisibility, of which we have 
 already studied two forms or grades, the molar and the mole- 
 cular ; the former consisting in the mechanical disruption of 
 large masses into small ones, the smallest still possessing sensible 
 magnitude ; while the latter is the further disruption, by physical 
 agents, such as heat, of moles or masses, whether large or small, 
 into their constituent molecules; that is to say, into parts contra- 
 distinguished from the minutest moles by the fact that they (the 
 said parts) possess no commensurable magnitude at all. In the 
 particular sample of matter which we selected for study, as being 
 the most familiar of all compounds, we saw molecular succeeding 
 to mere molar division, when heat melted comminuted ice into 
 water, and then raised water into invisible steam or gas, by cloth- 
 ing its molecules with the mutually repellent dynaspheres, each 
 dynasphere 1689 times larger than its material nucleus. 
 
 Infinitesimal as this subdivision of matter appears, inexpres- 
 sibly minute as we cannot but conceive the material particles to 
 be that form the central nuclei of the dynaspheres of bodies so 
 attenuated and rare as the invisible gases, we yet know we have 
 experimental proof that a further comminution of matter is 
 possible ; and that, as the smallest mass or mole of any compound 
 may be broken up into its constituent molecules, immeasurably 
 smaller still, so the ultimate molecule itself, however small we may 
 choose to conceive it, is nevertheless still a compound, consisting 
 of at least two parts, which, by chemical agency, may be detached 
 
150 FINAL TEEM OF THE DIVISIBILITY OF MATTER. 
 
 from each other, so as to resolve the compound into its ele- 
 ments. 
 
 Here the divisibility of matter, so far as our knowledge, and the 
 means of operating af our disposal, extend, reaches its final term. 
 The elementary bodies are, as we remember, so called precisely 
 because they resist every agency, mechanical, physical, and 
 chemical, which we can bring to bear in the hope of dividing or 
 decomposing them. We may imagine the two elementary par- 
 ticles which form the compound molecule of hydrochloric acid, 
 for example, to be as small as we please. In this respect we may 
 give the imagination free rein ; we may conceive the particle of 
 hydrogen, or of chlorine, to be divided and subdivided as many 
 millions of times as we like, or rather, until the imaginative 
 power is baffled by sheer exhaustion in the endeavour to push 
 this conception further. No experiment yet made tends to 
 restrict the freest range of our mental faculties in this direction ; 
 their only limitation lies in their own finite scope, doubtless more 
 or less extensive in different minds. But, when we have, each 
 of us, thus reached the idea of the smallest elementary particle 
 which it is within the power of the mind to picture, all experi- 
 ence stands opposed to our going still further, and presuming to 
 declare the elementary particles capable of division ad infinitum. 
 Not one experimental result can be adduced in support of such an 
 assertion. At this point, therefore, the experimental philosopher 
 arrests his inquiry. Beyond this limit he sees only the dream-land 
 of metaphysical speculation a region essentially sterile because 
 shut out from cultivation by means of experiment, from which alone 
 can spring the harvest of Truth in the proper sense of the word ; 
 having for its foundation natural facts ; for its object the study 
 of their relations ; for its result the determination of their laws. 
 
 To the metaphysical speculators, therefore, let us cheerfully 
 resign the utterly futile and fruitless discussion whether even 
 elementary matter may not be infinitely divisible. It is enough 
 for us to know that, at all events, we cannot infinitely divide it ; 
 but that, relatively to our powers and purposes, to the limits of 
 our imagination as well as of our experience, the assertion of 
 
THREEFOLD DIVISIBILITY OF MATTER. 151 
 
 the infinite divisibility of the elements is one we are not justi- 
 fied in making. 
 
 We thus arrive at the conception of indivisible particles as 
 the ultimate constituents of elementary bodies, and these par- 
 ticles have received the appropriate name of atoms (from the 
 Greek word re^vw, I cut, I divide, with the privative a prefixed 
 in token of negation). 
 
 The addition of this final term completes and enables us 
 to epitomise our view of the threefold divisibility of matter, 
 molar, molecular, and atomic ; the first (molar) being performed 
 by mechanical means, and resulting, when pushed to its utmost 
 limits, in masses or moles (clusters of molecules) characterized by 
 their possession of sensible magnitude ; the second (molecular) 
 accomplished by the agency of the physical forces (heat, electri- 
 city, &c.), employed under special conditions for the purpose, 
 and resulting in the production of the dynaspheric molecules of 
 which we reasonably conceive compound bodies to consist ; the 
 third (atomic) being capable of accomplishment only by agencies, 
 such, and so applied, as to produce chemical decomposition, 
 breaking up the incommensurable molecule itself into its ele- 
 mentary particles, which (as just explained) are called atoms, 
 because incapable of further disruption or comminution by any 
 means at our disposal. 
 
 This conception of the threefold divisibility of matter, molar, 
 molecular, and atomic, being once clearly understood, and firmly 
 grasped by the mind, we may usefully proceed, in the light which 
 this theory supplies, to compare as to their structure compound 
 with elementary gases. At first view we should be disposed, 
 perhaps, to anticipate as probable, that, while the compound gases 
 would be formed of divisible molecules or atom-clusters, the 
 elementary gases would present no such complexity of structure, 
 but consist merely of separate and indivisible elementary parti- 
 cles. But a little consideration will show us that this view is 
 incompatible with the results of our preceding inquiry. 
 
 Confining our attention, as usual, to the gases with which we 
 have made experimental acquaintance, let us select from among 
 
152 STRUCTURE OF MOLECULES, ELEMENTARY AND COMPOUND. 
 
 these, for the comparison in question, the simplest of our com- 
 pound gases, hydrochloric acid, and its two elementary con- 
 stituents, hydrogen and chlorine. And, that our views may be 
 the clearer on this subject, which we shall find to be important, 
 let us revert for a moment to our earliest experiments, and 
 refresh our memories by reviewing once again the volumetric 
 composition of hydrochloric acid gas. 
 
 Of this gas, we remember, the normal product-volume (2 litres) 
 is formed by the association of one unit-volume (1 litre) of 
 hydrogen with the same volume of chlorine gas ; their union 
 taking place without condensation. Each molecule of hydro- 
 chloric acid, therefore, is evidently constructed of two atoms at 
 least ; one being a hydrogen and the other a chlorine atom. 
 
 We have already gone over the experimental grounds for 
 believing that equal volumes of all gases, whether simple or 
 compound, contain, under like conditions, like numbers of dyna- 
 spheric molecules ; which are, therefore, necessarily of equal size 
 in all cases. 
 
 Let us now, to simplify our calculations, assign to the un- 
 known number n of hydrochloric molecules, existing in our 
 bilitral volume of hydrochloric acid gas, some definite numerical 
 value, say 1000. 
 
 This being assumed as the number of molecules in 2 litres, the 
 number in 1 litre is of course just half, or 500 ; and, as we re- 
 cognise that equal volumes of all gases contain equal numbers 
 of molecules, the litre of hydrogen and the litre of chlorine, 
 which go to the formation of our 2 litres of hydrochloric acid 
 gas, must likewise contain 500 molecules each. 
 
 Now, as each molecule of hydrochloric acid contains 1 atom of 
 hydrogen joined to 1 atom of chlorine, the 1000 molecules of 
 hydrochloric acid must, of necessity, contain 1000 atoms of 
 hydrogen joined to 1000 atoms of chlorine the whole number 
 of atoms present being therefore 2000. 
 
 But we have just seen that one litre of hydrogen and one litre 
 of chlorine contain, not 1000 molecules each of the respective 
 bodies, but only 500. 
 
DIATOMIC STRUCTUKE OF TYPICAL ELEMENTARY MOLECULES. 153 
 
 It follows clearly that 500 molecules of hydrogen and 500 
 molecules of chlorine have supplied respectively twice as many 
 atoms of those constituent bodies; each contributing its 1000 
 atoms to the aggregate number of 2000 atoms, existing in the 
 1000 hydrochloric molecules, contained in our 2 litres of hydro- 
 chloric acid gas. 
 
 If 500 molecules of an elementary gas supply 1000 atoms, 
 it is plain that each molecule supplies 2 atoms ; and thus we 
 clearly perceive that the molecule of the compound gas under 
 review, and the molecules of each of its elementary constituents, 
 are all formed on the same type that type being the first of 
 our quadruple series, viz., the hydrochloric acid or diatomic type. 
 
 This is a remarkable and striking, yet strictly logical deduc- 
 tion. It completes a chain of reasonings which, if correct, 
 justify the* conception that simple as well as compound gases 
 are complex as to their molecular structure ; and that this struc- 
 ture, for hydrogen and chlorine, is of the diatomic type, also 
 exemplified in hydrochloric acid. 
 
 Similar considerations show that this diatomic molecular struc- 
 ture characterises all the other permanent elementary gases with 
 which we have made acquaintance. 
 
 These newly-attained truths give rise to the necessity of a 
 new adaptation of our symbolic language for their expression. 
 Hitherto the symbols of the elements have merely expressed for 
 us their unit-volumes and combining weights ; and the formulae 
 of the compound gases have, in like manner, only represented to 
 us their product-volumes and combining weights. These ex- 
 pressions we have now to clothe with new meanings, corre- 
 sponding to our new conceptions concerning atoms and molecules 
 respectively. 
 
 To this new service we shall find our old formulae lend them- 
 selves with admirable flexibility. It is true, no doubt, that our old 
 expression for the unit-volume and combining weight of hydrogen 
 (H = 1) conveys no adequate picture of the diatomic structure 
 of its free molecule. But this symbol answers perfectly well to 
 express the single atom of hydrogen as it exists in combination 
 
 H 3 
 
154 DILITRAL OR MOLECULAR EXPRESSIONS 
 
 as, for example, we find it in the diatomic molecule (HC1) of 
 hydrochloric acid. Again, the expressions employed to repre- 
 sent the product-volumes and weights of the compound gases 
 require no modification to fit them for representing the atomic 
 structure of the molecules of those gases. For example, the 
 formula, 
 
 H + 01 = HC1 = 1 + 35-5 = 36-5, 
 
 affords as perfect a picture of the diatomic molecule of hydro- 
 chloric acid as it does of its bivolumetric composition ; at the 
 same time accurately expressing the weight of its product- 
 volume. And lastly, as we know that the molecules or atom- 
 clusters of the free elementary gases are formed on the binary 
 type of hydrochloric acid, we may evidently seek in the formula 
 of that acid a model on which to frame appropriate symbolic 
 expressions for those free elementary molecules. 
 
 Thus, for example, the molecular formulae for free hydrogen 
 and free chlorine are respectively 
 
 for hydrogen H + H = HH =1+1=2, 
 and for chlorine Cl + Cl = C1C1 = 35'5 + 35-5 = 71. 
 
 Hence, it further appears that, for the volumetric symboliza- 
 tion of the free elementary gases, our formerly-used monolitral 
 (and, in exceptional cases, hemilitral) expressions are no longer 
 appropriate, but must be replaced by dilitral symbols like 
 those employed to denote the normal product-volumes of the 
 compound gases. 
 
 We thus obtain, as the correct molecular expressions for 
 hydrochloric acid and its elements, the following bi-volumetric 
 symbols : 
 
 HC1 
 HH 
 
 C1C1 
 
CONTRASTED WITH MONOLITBAL OR ATOMIC EXPRESSIONS. 155 
 
 Upon these models the formulae of the other elementary gases 
 may be readily framed ; and the equations representing the 
 reactions in which they participate must of course undergo 
 corresponding modifications. For example, the expression 
 H + Cl = HC1, or, diagrammatically represented, 
 
 heretofore employed to represent the synthesis of hydrochloric 
 acid by the direct combination of the two free gases, its consti- 
 tuents, is no longer, for us, a true representation of the facts : 
 seeing that we know those two gases to be formed not, as from 
 this equation would appear, by the reaction of solitary atoms, but 
 by that of diatomic molecules. The true molecular equation of 
 this synthesis becomes, therefore, 
 
 HH 4. C1C1 = HC1 + HC1, 
 or, diagrammatically represented, 
 
 HH 
 
 or, again, reduced to its most succinct expression, 
 
 2H + 2C1 = 2HC1. 
 
 The dilitral symbolization of the elementary gases has a 
 further advantage; it brings them into direct comparison, 
 volume for volume, with the compound gases (also dilitral, as 
 we remember), so that a list of the molecular weights of the 
 elementary and compound gases represents also their relative, 
 volume-weights or specific gravities. We may express these either 
 in terms of the unit- vol. of hydrogen, 
 
 [H] = i; 
 
 or, preferably, in terms of the product-volume of hydrogen, 
 
156 
 
 TABLE OF MONOLITRAL AND DILITRAL FORMULAE. 
 
 and we thus obtain the two symbolic and numerical series 
 which are placed side by side in the following table : 
 
 
 SPECIFIC GRAVITIES. 
 
 -Names or 
 gases, 
 
 Monolitral. 
 
 
 Dilitral. 
 
 elementary and 
 
 
 
 
 
 compound. 
 
 I 
 
 I = 1 
 
 H 
 
 H =2 
 
 Hydrogen . . 
 
 H = 
 
 1 
 
 HH = 1 
 
 X 2 = 2 
 
 Chlorine . . 
 
 Cl = 
 
 35-5 
 
 C1C1 = 35-5 x 2 = 71 
 
 Oxygen 
 
 = 
 
 16 
 
 OO = 16 
 
 X 2 = 32 
 
 Nitrogen . . 
 
 N = 
 
 14 
 
 NN - 14 
 
 X 2 = 28 
 
 Hydrochloric acid 
 
 HC1 
 
 36-5 -j p. OK 
 
 TTP1 
 
 36-5 
 
 2 
 
 2 
 
 
 Water-gas . . 
 
 H 2 O 
 
 IT = 9 
 
 H 2 = .. 
 
 18 
 
 Ammonia . . . . 
 
 H 3 N 
 
 2 
 
 ~ = 8-5 
 
 H 3 N . .. 
 
 17 
 
 M!arsh-gas 
 
 H 4 C 
 
 2 
 
 - 
 
 H 4 C = .. 
 
 16 
 
 It will be observed that, in the monolitral column of this 
 table, the formulae of the compound gases are halved, while in 
 the dilitral column the symbols of the elementary gases arc 
 doubled, in order to render the whole series comparable in 
 terms of the standard adopted in each case. Our preference is 
 given to the second arrangement, because it distinctly shadows 
 forth the true molecular structure of the free elementary gases, 
 depicting clearly their analogy in this respect with the molecular 
 type of hydrochloric acid. And, while this important chemical 
 conception is figured in the four first expressions of the dilitral 
 column, its four last members remind us of another and an 
 equally significant chemical truth, viz., that the elementary 
 gases, though in the free state they always exist as molecules, 
 very commonly split up into separate atoms when they enter into 
 combination. Illustrations of this fact, with reference to hydro- 
 
(XXNTRADISTINCTIVE DEFINITIONS OF ATOM AND MOLECULE. 157 
 
 gen, are supplied by the expressions HC1 and H 3 N in the table. 
 For, though we might conceive the other two expressions, H 2 O 
 and H 4 C to represent, the former the union of one hydrogen- 
 molecule (HH), and the latter of two hydrogen-molecules (2HH) 
 with oxygen and carbon respectively, we cannot but see in 
 the formula HC1 the representation of a single hydrogen 
 atom (H - 1) associated with a single chlorine atom (Cl = 35'5) ; 
 while, in the formula H 3 N, we are equally constrained to recog- 
 nize the representation of at least one separate hydrogen atom, 
 namely the third, that one which, in the expression HHH, is 
 in excess of the diatomic expression HH, corresponding to the 
 molecule of free hydrogen. 
 
 This consideration is of value, because it enables us to assign 
 to the molecules and atoms of the elementary bodies contra- 
 distinctive definitions, quite irrespectively of any question as to 
 the greater or less divisibility of matter. That is to say, we may 
 define the atom of any given elementary body to be the smallest 
 proportional weight thereof which is capable of existing in chemi- 
 cal combination ; and we may define the molecule of an elementary 
 body to be the smallest proportional weight thereof which is 
 capable of existing in the free or uncombined state. 
 
 This definition of the elementary molecule, it will be observed, 
 does not assume for it any particular numerical relation to the 
 elementary atom ; its terms are wide enough to admit the con- 
 ception not only of diatomic, but also of triatomic, tetratomic, 
 and polyatomic molecules ; while, on the other hand, its terms 
 do not exclude the conception of coincidence in weight between 
 the atom, or combining minim, and the molecule, or free minim, of 
 an elementary body; in other words, this definition admits as 
 possible the conception of a monatomic molecule. 
 
 With the above models of molecular formulas before us we 
 shall have little difficulty in extending this mode of symbolization 
 to the other elements which we have passed in review, viz., to 
 the analogues of chlorine, bromine and iodine ; to the analogues 
 of oxygen, sulphur and selenium ; to the analogues of nitrogen, 
 phosphorus and arsenic ; and lastly, to the analogues of carbon, 
 
158 MOLECULAR SYMBOLS OF TYPICAL ELEMENTS & THEIR CONGENERS. 
 
 silicon and titanium. In the cases of phosphorus and arsenic, 
 we shall, of course, again encounter the same exceptional volu- 
 metric relations that obliged us to modify our former symboli- 
 zation of these bodies ; and we shall have to meet the anomaly 
 by a corresponding modification of the molecular expressions. 
 The following table presents a collective view of these elements, 
 and of their molecular construction and symbolization : 
 
 MOLECULAR CONSTRUCTION AND SYMBOLIZATION OF ELEMENTARY BODIES : 
 Chlorine Group. 
 
EXCEPTIONAL MOLECULES OF PHOSPHORUS AND ARSENIC. 159 
 
 Carbon Group. 
 
 Si 
 
 Ti 
 
 C 
 
 | Si 
 
 I 1 
 
 CC = 24 (?) 
 
 ... -A J 
 
 SiSi 
 
 = 57 (?) 
 
 TiTi - 100 (?) 
 
 The symbolization of PPPP and AsAsAsAs in this table 
 explains itself at a glance. The dilitral weight of these two 
 bodies, that is to say, the weight of their molecule, or free 
 minim, in the state of vapour, is four times as great as that of 
 their atom, or combining minim, as determined in phosphoretted 
 hydrogen, H 3 P, and arsenetted hydrogen, H 3 As. This, indeed, 
 follows as a corollary from our previous observations on these 
 bodies ; for, as their unit volume, or monolitral volume, had to 
 be represented respectively by PP, or 2P, and AsAs, or 2As, 
 it is obvious that their dilitral or molecular formula must be 
 twice as great; or 2P x 2 = 4P, and 2 As x 2 = 4 As. In these 
 cases, accordingly, the molecule is tetratomic. In the carbon 
 group we adhere, it will be observed, to the use of dotted lines 
 in the diagrammatic symbols to imply that the expressions are 
 hypothetical ; and we add a note of interrogation within brackets 
 to the written symbols for the same purpose. 
 
 With the molecular equation of the synthesis of hydrochloric 
 acid as a model, we may readily reconstruct, in the molecular 
 form, our old atomic equations representing the syntheses of 
 other compound gases. It will suffice, therefore, to give one or 
 two examples of such syntheses, expressed in molecular formulas, 
 and for this purpose we will select water and ammonia. Our 
 old formula of the formation of water from its elements was 
 
 for which we now substitute 
 2HH 
 
 = 2H 2 0. 
 
160 COMPARISON OF ATOMIC AND MOLECULAK EQUATIONS. 
 
 Similarly, should means be found to form ammonia directly 
 from its elements, the atomic equation representing its syn- 
 thesis would be 
 
 3H + N = H 3 N; 
 
 for which, substituting the molecular expression, we have 
 3HH + NN = 2H 3 N. 
 
 Once more : just as synthetic equations, adapted to the mole- 
 cular hypothesis, exhibit, as starting-points of the reactions, 
 molecules, not atoms, of the elements taking part therein ; just 
 so, conversely, molecular equations, when employed to represent 
 reactions of decomposition, must be so constructed as to express 
 in molecules, not in atoms, the elements set free. 
 
 Thus, the ordinary equations which, in a previous lecture, 
 represented for us the decomposition of hydrochloric acid, water, 
 and ammonia, by sodium (Na), were as follows : 
 
 HC1 + Na NaCl + H. 
 H 2 4- 2Na - Na 2 + 2H. 
 H 3 N + 3Na - Na 3 N + 3H. 
 
 The molecular equations representing these decompositions 
 are 
 
 2HC1 + NaNa = 2Na Cl + HH. 
 
 H 2 + NaNa - Na 2 O + HH. 
 
 2H 3 N + SNaNa - 2Na 3 N + 3HH. 
 
 In the former of these two series of formulae it will be noticed 
 that the first and third equations represent Na as used, and 
 H as escaping, wholly or partly, in single atoms. In the cor- 
 responding equations of the second group, the expressions 
 employed for free sodium and hydrogen represent them as con- 
 sisting of atoms in couplets, i.e., as diatomic molecules ; a struc- 
 ture to whose type we know that hydrogen conforms, while, 
 strong analogies, justify us in assuming a similar construction for 
 free sodium gas. On comparing with each other the middle 
 
MOLECULAR FORMULAE OF THE NITROXYGEN SERIES. 161 
 
 equation of each set, it will be seen that they are identical 
 expressions ; a circumstance due to the fact that, in the decom- 
 position of water by sodium, the proportion both of the metal 
 employed and of the hydrogen set free is 1 molecule = 2 
 atoms (comp. p. 218). 
 
 The two following series of formulae set in like contrast the 
 old and new modes of symbolizing the decomposition of water 
 and ammonia by chlorine : 
 
 Atomic Notation. 
 
 H,0 + 2C1 = 2HC1 + 0. 
 H 3 N + 3C1 - 3HC1 + N. 
 
 Molecular Notation. 
 
 2H,O + 2C1C1 = 4HC1 + 00. 
 2H 3 N + 3CIC1 = 6HC1 + NN. 
 
 In constructing molecular equations of compound bodies care 
 must be taken to select those expressions which, while repre- 
 senting the true relative proportions of the elementary consti- 
 tuents, embody for this purpose the smallest number of atoms 
 with which the compound or product-molecule can be built up. 
 On the other hand, symbolic expressions must never comprise 
 fractional parts of atoms : such fractional formulae are of course 
 inadmissible, implying, as they would do, the division of that 
 which, by hypothesis, and by the name founded thereon, is recog- 
 nized as indivisible. The rules here laid down are well exempli- 
 fied in the following table of the molecular formulae, representing 
 the oxides of nitrogen : 
 
 Molecular formulae, of the Nitroxygen series. 
 
 1 mol. of nitrous oxide = 2 at. of nitrogen + 1 at. of oxygen = N 2 O. 
 
 1 mol. of nitric oxide = 1 at. +1 at. = N O. 
 
 1 mol. of nitrous acid = 2 at. ' +3 at. = X 2 O 3 . 
 
 1 mol. of hyponitric acid = 1 at. +2 at. = N O 2 . 
 
 1 mol. of nitric acid = 2 at. +5 at. = N 2 O 3 . 
 
162 SPECIAL APTITUDES OF ATOMIC AND MOLECULAR NOTATIONS. 
 
 The value of this conception of the molecular structure of 
 elementary as well as compound gases will become more apparent 
 to us in future stages of our inquiry than it is at present. But 
 the distinctness which the atomic theory lends to our views of 
 material phenomena can be immediately perceived. It affords 
 us a satisfactory explanation of the definiteness and immutability 
 of chemical composition ; of the step-by-step gradations, in 
 simple multiple ratios, by which the proportions of such elements 
 as form more than one compound with each other are observed 
 to vary ; and it enables us readily to understand the fact that 
 compounds in so many cases become less stable as they increase 
 in complexity ; in other words, as their molecules are built up 
 of a larger number of atoms. 
 
 The incorporation of these views in the formulas of simple and 
 compound bodies, and of their reactions, evidently impresses 
 upon our symbolic language a new significance, and adapts it to 
 aid in theoretically interpreting the phenomena which it also 
 depicts and records. 
 
 We have only to add on this subject that both the atomic and 
 molecular forms of symbolic expression have their peculiar 
 merit ; the former being more succinct, the latter more compre- 
 hensive. When very complete and encyclopaedic expressions 
 are required, including the proportions, both by volume and by 
 weight, as well of the bodies brought into action as of the 
 resultant products, simple and compound, molecular formulae are 
 indispensable. When only the relative weights of the bodies 
 in action, and of their products, require representation (as in the 
 majority of practical problems), atomic equations are sufficiently 
 comprehensive, and have the advantage in point of conciseness 
 and simplicity. Many chemists indeed use this latter form 
 only of the symbolic language ; let it be our care to master both 
 modes of expression, so that we may be able to employ each in 
 turn for its appropriate purposes. 
 
163 
 
 LECTURE X. 
 
 Molecular arid atomic constitution of the typical compounds curious rela- 
 tions, ponderal, numerical, and potential, of the typical elementary atoms 
 two sorts of chemical value or power, molecule-forming and atom-fixing 
 unitary standard of atom-fixing power major and minor equivalent weights 
 coefficients of atom-fixing power, or quantivalence comparative quan- 
 tivalence of the typical elements and their congeners germ of a natural 
 system of chemical classification volume-condensing power of atoms, how 
 far proportionate to their quantivalence alternative standards of quanti- 
 valence chemical value in exchange exemplification thereof in the 
 syntheses of hydrochloric acid and water, and in the hypothetical synthe- 
 ses of ammonia and marsh-gas also in their decomposition by chlorine 
 in the contrasted action of chlorine and oxygen on hydriodic acid and in 
 the comparative structure of ammonia and nitrous acid quantivalential 
 equilibrium of the nitroxygen series distinction between numerical and 
 potential quantivalence also between quantivalence and chemism 
 transitional state of the question tabular summary. 
 
 IN the light of our new conceptions concerning the molecules 
 and atoms of which the elementary bodies and their compounds 
 are built up, we reviewed, at our last meeting, the diagrammatic 
 symbols previously employed to represent the volumetric and 
 ponderal composition of hydrochloric acid, water, ammonia, and 
 marsh-gas ; and we found that the double squares, expressing 
 the dilitral product-volumes of these compounds, are perfectly 
 well adapted to represent for us their respective free molecules ; 
 while the single squares, which previously served us to denote 
 the monolitral unit-volumes of their respective constituents, 
 answer equally well to depict the combining atoms of those ele- 
 mentary bodies. 
 
 In the following diagram the dilitral product-volumes, or, as 
 we must now say, the molecules, of our four typical compounds, 
 are placed in a column by themselves, in contrast with the 
 monolitral unit-volumes, or, in our present view, the atoms of 
 the elements which they respectively contain ; the arrangement 
 
164 
 
 TYPES OF ATOMIC AND MOLECULAR CONSTITUTION. 
 
 being such that the atoms of chlorine, oxygen, nitrogen, and 
 carbon, occupy the second column of the diagram, with the 
 hydrogen-atoms they respectively take up displayed on the right- 
 hand side, and the resulting compound molecules on the left. 
 
 MOLECULAR AND ATOMIC CONSTITUTION OF THE FOUR 
 TYPICAL COMPOUNDS. 
 
 Product-volumes 
 = Molecules. 
 
 Unit-volumes 
 = Atoms. 
 
 H 
 
 
 H 
 
 H 
 
 
 H 
 
 
 H 
 
 C12; 
 
 H 
 
 
 H 
 
 
 H 
 
 
 H 
 
 In examining this diagram, we are at once struck with the 
 fact that the four elements, displayed in the second column, stand 
 very differently related, on the one hand, to the volume and weight 
 of the compound molecules they respectively form, and, on the 
 other hand, to the volume and number of the atoms that take 
 part with them in forming those molecules. 
 
 Thus, a glance at the left and central columns shows us that, 
 though the volumes of the compound molecules are equal, they 
 contain very unequal weights of the four elements under con- 
 sideration, viz., 35-5 of chlorine, 16 of oxygen, 14 of nitrogen, 
 and 12 of carbon respectively. 
 
 Again, looking to the number of hydrogen-atoms depicted on 
 the right-hand side, we see that the atoms of the four central 
 elements, chlorine, oxygen, nitrogen, and carbon, stand related 
 respectively to 1, 2, 3, and 4 hydrogen-atoms. 
 
THEIR CURIOUS NUMERICAL RELATIONS. 165 
 
 It is a curious circumstance, and one which the diagram 
 almost forces us to notice, that the heaviest of the four central 
 atoms (Cl = 35'5) is precisely the one which engages the small- 
 est number of hydrogen-atoms, viz., only one; while the other 
 three (0 = 16, N = 14, and C = 12), as they grow successively 
 lighter, engage increasing numbers of hydrogen-atoms, viz., 2, 
 3, and 4 atoms respectively. 
 
 In other words, it takes the whole atom-power of chlorine, 
 35'5, to engage 1 atom of hydrogen ; whereas, the atom- power of 
 oxygen, 16, suffices to engage 2 hydrogen-atoms ; and the atom- 
 power of nitrogen and carbon suffice, respectively, to engage 3 
 and 4 hydrogen-atoms. 
 
 It is impossible to overlook the singular numerical relations 
 which these unequal atom-engaging powers bring about in the 
 product-volumes, or molecules, depicted on the left side of the 
 diagram. While carbon, nitrogen, and oxygen have, as their 
 atom weights, 12, 14, and 16, respectively, the molecules they 
 form, in combining with hydrogen, weigh 16, 17, and 18 respec- 
 tively ; the molecular weights advancing by 1 only, while the 
 atomic weights advance by 2, at each grade. This is made still 
 more remarkable by the fact that from 18, the weight of the 
 water-gas molecule, to 36*5, the weight of the hydrochloric acid 
 molecule, the advance is by a sudden spring to about double the 
 first-named quantity ; 18 x 2 being 36, while the actual weight 
 of the hydrochloric acid molecule is 36-5. Another element 
 (fluorine), to which our attention has not been directed, fills up, 
 in an equally interesting manner, an intermediate link in this 
 numerical chain, as we shall hereafter learn. These curious rela- 
 tions are entirely unexplained, though they have latterly attracted 
 much attention. They do not, however, belong to our present 
 inquiry ; from which, it must be owned, we have digressed for a 
 moment to bestow on them this passing notice. 
 
 Eeturning to our immediate study, we observe that the table 
 places before us the four centrally-disposed elements, in two per- 
 fectly distinct chemical relations ; the first more especially 
 volumetric and molecular, the second essentially numerical and 
 
166 CHEMICAL VALUES, MOLECULE-FORMING AND ATOM-FIXING. 
 
 atomic. Hence, two parallel series of minimum- weights ; one 
 representing the minimum quantity of each element requisite to 
 take part in the formation of a compound molecule ; the other 
 corresponding to the minimum quantity of each element which 
 is adequate to engage or fix one standard atom. 
 
 As the standard atom, whereby to measure this atom-fixing 
 power of the elements, the table presents us with hydrogen 
 (H = 1), and this unit we gladly adopt : first, because hydrogen 
 is also our unitary standard for relative volume-weight or specific 
 gravity ; and for concrete volume-weight (i. e., specific gravity 
 read in criths) ; secondly, because hydrogen is, as we shall pre- 
 sently see, particularly well suited to serve as our unit for this 
 additional purpose. 
 
 Working out the comparison of these two sorts of chemical 
 value, the molecule-forming and the atom-fixing, for hydrogen 
 and the four other elementary bodies under review, we obtain 
 the three following columns of figures : 
 
 CHEMICAL VALUES, molecule-forming, and atom-fixing, of the 
 standard element, hydrogen, and of the four typical elements ; 
 with the ratios of those values. 
 
 The four Typical Elements, pre- 
 
 Minimum weights thereof 
 
 
 ceded by the Standard Element, 
 Hydrogen. 
 
 requisite 
 
 Ratios of the num- 
 bers in Columns 
 3 and 4. 
 
 To take part in 
 
 To engage 
 
 
 
 Their names. 
 
 Their literal 
 symbols. 
 
 the formation 
 of a molecule. 
 
 one 
 standard atom. 
 
 
 1. 
 
 Q 
 
 3. 
 
 4. 
 
 5. 
 
 Hydrogen 
 
 H 
 
 1 
 
 1 
 
 1 j 
 
 1 
 
 Chlorine . . 
 
 Cl 
 
 35-5 
 
 35-5 
 
 35-5 
 
 Oxygen . . 
 
 
 
 16 
 
 8 
 
 "lY 
 
 T 
 
 Nitrogen . . 
 
 N 
 
 14 
 
 4-66 
 
 __ Q 
 
 4-66 
 
 
 
 
 
 12 
 
 Carbon . . 
 
 C 
 
 12 
 
 3 
 
 ~3~ = ' 
 
MAJOR AND MINOE EQUIVALENT WEIGHTS. 167 
 
 In this table, it will be observed, column 3 represents the 
 molecule-forming equivalents of the elements, or the proportions 
 by weight in which they can replace each other in contributing 
 to the construction of a molecule; while column 4 sets forth 
 the atom-fixing equivalents of the elements, or the proportions 
 in which they can replace each other in fixing a standard atom. 
 Column 5 shows the ratios of the molecule-forming to the atom- 
 fixing weights ; and the usefulness of these ratios we will now 
 proceed to consider. 
 
 Taking, for example, the last element in the table, carbon, we 
 see that its molecule-forming minimum- weight is 12, while its 
 standard -atom-fixing minimum- weight is 3 ; and, on referring to 
 the diagrammatic formula of marsh-gas (p. 106) we find that, in 
 reality, 12 parts by weight of carbon enter into the construction 
 of the marsh-gas molecule, while these 12 parts fix, in that mole- 
 cule, 4 atoms of hydrogen ; each of which, therefore, is fixed by 
 i-j.2 = 3 parts by weight of carbon. 
 
 So likewise, of nitrogen (the element next above carbon in 
 the table), we see that 14 is the molecule-forming minimum, and 
 * = 4' 6 6 the atom-fixing minimum ; of oxygen, these two 
 minima are respectively 16 and l -f =8; while, for chlorine, 
 which fixes only 1 standard atom, the two minima of course 
 coincide. 
 
 As for hydrogen, it stands apart, at the head of each column, 
 the unitary standard of atom-fixing power. 
 
 From these considerations it is clear that we might attach to 
 each element two representative or equivalent numbers ; one 
 expressing its minimum-weight relatively to the formation of a 
 molecule, the other its minimum-weight relatively to the fixa- 
 tion of an atom ; and we might distinguish these as its major 
 and minor equivalent weights, or by some other distinctive desig- 
 nations. 
 
 It would, however, be obviously inconvenient thus to have 
 in use two sets of minima weights, or equivalents. Such a 
 duplicate system of notation would encumber the memory, and 
 greatly impair the succinctness of our symbolic short-hand. In 
 
168 COEFFICIENTS OF ATOM-FIXING POWER OR QUANTIVALENCE. 
 
 order to avoid these evils it is desirable to include, for each ele- 
 ment, the two separate weights in a single concise expression. 
 
 For this purpose it suffices to attach, to each of the molecule- 
 forming minimum-weights, given in the third column of the 
 table, a coefficient of atom-fixing power ; that is to say, a sign 
 expressing how many standard atoms its said weight is adequate 
 to satisfy. This is readily done by aid of the ratios set forth in 
 column 5 of the table. These ratios are, in fact, the coefficients 
 in question ; and by writing them (in Eoman numerals for dis- 
 tinctness' sake) over against the molecule-forming minimum- 
 weights, in the ordinary place of exponents, we learn at a glance 
 the number of standard atoms which the said weights can respec- 
 tively fix. 
 
 Accordingly, the molecule-forming minimum of chlorine = 
 35*5 is written 35*5* ; the corresponding minimum of oxygen 
 16 is written 16 n ; that of nitrogen = 14 is written 14 m ; 
 and that of carbon = 12 is written 12. Or, for still greater 
 brevity, as the signs Cl, 0, N, and C, are already associated in 
 our memory with their respective molecule-forming weights, we 
 may attach the coefficients directly to these symbols, writing 
 them respectively : 
 
 Cl 1 , O n , N ra , and C". 
 
 Dashes, one, two, three, or four in number, are commonly em- 
 ployed by chemists instead of numerals to express these coeffi- 
 cients ; and they answer the purpose equally well for the lower 
 expressions. But, for numbers higher than three, the numerical 
 expressions are preferable to the dashes, as being easier than they, 
 both to write and read ; and, to preserve uniformity throughout, 
 I prefer to use Eoman numerals for these signs in all cases. 
 
 We are in want of a good appellation to denote this atom- 
 fixing power of the elements. The vague and rather barbarous ex- 
 pression, atomicity, has drifted into use for this purpose ; and the 
 elements have been called monatomic, cfo'atomic, In'atomic, and tetra- 
 tomic, accordingly as their respective molecule-forming minimum- 
 weights are capable of saturating 1, 2, 3, or 4 standard atoms. 
 
QUANTIVALENTIAL CLASSIFICATION OF CHEMICAL ELEMENTS. 169 
 
 These expressions are faulty, because they are open to misinter- 
 pretation, as if intended to denote the atomic structure of the 
 respective elementary molecules themselves ; a sort of confusion, 
 the possibility of which should always be sedulously avoided in 
 scientific nomenclature. 
 
 We shall escape this evil by substituting the expression 
 quantivalence for atomicity ; and designating the elements uni- 
 valent, frz'valent, /n'valent, and quadrivalent, according to their 
 respective atom-fixing values. 
 
 As to their molecule-forming values, these may be indifferently 
 termed their atom-weights, or combining numbers ; which, it 
 will be remembered, correspond, for the volatile elements, with 
 a few exceptions only, to their respective gas-volume-weights, or 
 specific gravities relatively to hydrogen. 
 
 However denoted, the two sorts of chemical value remain in 
 themselves most clearly distinguished ; and the importance of 
 keeping this distinction in mind cannot be overrated. By the 
 difference of their quantimlent powers the four elements under 
 review are impressed, each with a strongly-marked character of 
 its own ; and we shall presently see that each of the four stands 
 in this respect at the head of a group of congeners, endowed with 
 like atomic quantivalence ; so that we have here the first germ 
 of a natural classification of chemical elements, based on experi- 
 ment, and conformable with truth. 
 
 Thus, for example, to univalent chlorine (CP) correspond 
 bromine and iodine, both likewise univalent, and written accord- 
 ingly Br 1 and I 1 . 
 
 So, again, to bivalent oxygen (O n ) correspond bivalent sulphur 
 and selenium S n and Se n respectively. 
 
 Trivalent nitrogen (N ra ) has for its congeners trivalent phos- 
 phorus and arsenic, P ra and As. 
 
 Lastly, quadrivalent carbon (C 1 *) is the typical head of the 
 group which comprises quadrivalent silicon and titanium, 
 written Si^ and Ti iv . 
 
 It should not be overlooked that, with the varied atom-fixing 
 or quantivalent powers are coupled, in the case of the typical 
 
 i 
 
170 RELATION OF QUANTIVALENCE TO CONDENSING POWER. 
 
 elements, and of their respective combinations with hydrogen, 
 proportionate volume-condensing powers ; these two powers (as 
 we remember) going together, and increasing pari passu. Thus, 
 the four atoms of hydrogen fixed by quadrivalent carbon in 
 marsh-gas are not only four times as numerous, but also four 
 times as compactly disposed, as the one hydrogen-atom fixed by 
 univalent chlorine in hydrochloric acid gas. To this rule, 
 however, phosphorus and arsenic are, to a certain extent, excep- 
 tions. These trivalent bodies do, indeed, fix as many hydrogen- 
 atoms as their trivalent architype, nitrogen ; and they fix them, 
 moreover, within the same dilitral product-volume. Yet, re- 
 membering as we do that the normal nitrogen-atom occupies 
 twice as much space as the exceptional atoms of phosphorus and 
 arsenic respectively, we are fain to admit that the space left for 
 hydrogen-atoms in the respective product-volumes of phosphor- 
 etted and arsenetted hydrogen is greater, by exactly half a volume, 
 than the space left for hydrogen-atoms in the product-volume of 
 ammonia. In other words, the total condensation in the case of 
 ammonia is represented by the ratio f = ^; whereas the total 
 condensation, in the cases of phosphoretted and arsenetted 
 hydrogen, is only in the ratio jr~ = -!L As for carbon, its free 
 vapour having never been obtained, we have no means of deter- 
 mining the space filled by its atom in gaseous compounds. It 
 is only therefore provisionally, and on analogical, not experi- 
 mental, grounds, that we place it volumetrically on a par with 
 the other elements under consideration. Hence the dotted lines 
 with which (as before) the carbon-atom is represented in our 
 diagram. 
 
 How far the coincidence between the atom-fixing and volume- 
 condensing powers of the typical elements observed in their 
 combinations with hydrogen, may extend to their combina- 
 tions with other bodies, and how far such coincidence may be 
 traceable in the combining relations of the elements generally, 
 these are problems as yet unsolved. 
 
 Reverting to the simple consideration of quantivalence itself, 
 we of course understand that the expressions uni, bi, tri, and 
 
CHEMICAL VALUE IN EXCHANGE. 171 
 
 quadrivalent would have to be modified, if the atom selected as 
 a unitary standard of quantivalence were itself of other than 
 univalent power. For example, were the bivalent oxygen-atom 
 selected, as the standard or unit of quantivalence, hydrogen, 
 having only half that value, would have to be called semivalent ; 
 while nitrogen, having once and a half that value, would be 
 sesquivalent ; and carbon, having twice that value, would be bi- 
 valent. Thus, though the degrees of quantivalence would be 
 quite as correctly expressed in terms of a bivalent as of a univa- 
 lent standard, two expressions out of the four would be fractional 
 a manifest inconvenience. This inconvenience would become 
 increased the fractional expressions would become more nume- 
 rous, were a trivalent or quadrivalent atom, such as the atom 
 of nitrogen, or that of carbon, adopted as the standard of quan- 
 tivalence. If, therefore, we had not already freely chosen uni- 
 valent hydrogen as our standard of quantivalence, we should 
 have been constrained, by the exhaustive method, to its adoption, 
 as the simplest and best for the purpose 
 
 The unequal molecule-forming powers of the elementary bodies, 
 represented by the different weights of their atoms, and their 
 unequal atom-fixing powers, represented by their dissimilar co- 
 efficients of quantivalence, show us that each of these bodies 
 possesses what may be termed its specific chemical value in 
 exchange. 
 
 Thus, with respect to the power of forming a molecule, we 
 know already, and are again reminded by looking at our last 
 diagram, that 12 parts by weight of carbon are " worth " as much 
 for this purpose as 14 parts of nitrogen, 16 of oxygen, and no 
 less than 35'5 of chlorine. 
 
 So again, with reference to the power of fixing a standard 
 atom, the elements comprised in each of the four groups whereof 
 chlorine, oxygen, nitrogen, and carbon are the respective types, 
 possess for this purpose chemical value in exchange, varying by 
 gradations, as we have just seen, from 1, for the first group, 
 through 2 and 3 for the second and third, up to 4 for the last. 
 
 In other words, and to borrow a financial mode of expression, 
 
 i2 
 
172 RELATION OF ATOM-FIXATION TO ATOM-DISPLACEMENT, 
 
 one atom of any element in group 4 (the quadrivalent group) is 
 exchangeable at par far four atoms of any element in group 1 (the 
 univalent group), and for two atoms of any element in group 
 2 (the bivalent group). So also, as to the reciprocal exchange- 
 ability at par of the atoms of elements comprised respectively in 
 the trivalent and quadrivalent groups, the simplest way to repre- 
 sent this equation, without resorting to fractional expressions, is 
 to describe three atoms of any of the quadrivalent elements as 
 " worth," for atom-fixing purposes, four atoms of either of the 
 trivalent bodies. 
 
 From this conception of "value in exchange" we readily 
 gather that the atomic relations which we designate quaiiti- 
 valence imply not only atom-fixing, but also atom-displacing 
 power, so that, in learning how many standard units of 
 quantivalence any given elementary atom can attract and re- 
 tain within a compound molecule, we learn also how many it 
 can remove therefrom, when it is employed as a decomposing 
 agent, under conditions enabling it to eliminate, partly or 
 wholly, one or more of the constituents thereof. 
 
 The differences between the elements, as to their chemical 
 value in exchange, for molecule-forming and atom-fixing (or free- 
 ing) purposes, are well illustrated in the simple equations which 
 represent the syntheses of hydrochloric acid and water respec- 
 tively. 
 
 We referred to these equations when speaking of the mole- 
 cular structure of the typical elementary gases; and we then 
 pointed out that, when two such gases combine directly to form 
 a compound gas, there is a reciprocal exchange of atoms between 
 their diatomic molecules ; which exchange, we may now add, 
 bespeaks, by its conditions, their relative molecule-forming and 
 atom-fixing powers. 
 
 Contrast, for example, the equation representing the synthesis 
 of hydrogen and chlorine (both univalent elements) to form 
 hydrochloric acid, with the equation representing the synthesis 
 of hydrogen and oxygen (the latter a bivalent element) to form 
 water. 
 
AS DISPLAYED IN SYNTHESES OF HYDROCHLORIC ACID AND WATER. 173 
 
 In the first case we have 
 
 Synthesis of Hydrochloric Acid. 
 
 Constituent Gases. Hydrochloric Acid Gas produced. 
 
 OrHH + CrCP= 2 HC1 1 
 
 Each atom of univalent chlorine here passing in exchange, at 
 par, for one atom of univalent hydrogen ; so that we may regard 
 hydrochloric acid (HC1) as a molecule of hydrogen (HH) in 
 which one atom of hydrogen (H) is replaced by one atom of 
 chlorine (Cl), (or vice versa). 
 
 In the second case we have a very diiferent expression ; for 
 here bivalent oxygen comes into play, and we accordingly obtain 
 the equation : 
 
 Synthesis of Water. 
 
 Constituent Gases. 
 
 2 
 
 7< 
 / 
 
 eir distribution to 
 arm Water-gas. 
 
 Th 
 
 i 
 
 eir condensation 
 nto Water-gas. 
 
 HH 
 
 + 
 
 
 H 
 
 
 
 + - 
 
 
 H 
 
 H 2 n 
 
 
 IT 0" 
 
 O n 
 
 
 
 
 
 o n = 
 
 
 HH 
 
 H 
 
 H 
 
 H 2 n 
 
 
 
 
 Or 2HH + O n O n - 2H 2 O n . 
 
 In the water-gas molecule we see bivalent oxygen fixing twice 
 as many atoms of hydrogen as univalent chlorine fixes in hydro- 
 chloric acid ; and the equation shows us that, in the act of com- 
 bination, two standard univalent hydrogen-atoms are, so to speak, 
 bartered, or given in exchange at par, for one atom of divalent 
 oxygen. 
 
 In the two following diagrams we have represented, in like 
 manner, for the further illustration of our subject, the (as yet 
 
174 ALSO IN HYPOTHETICAL SYNTHESES OF AMMONIA AND MAKSH-GAS. 
 
 theoretical) syntheses of ammonia and marsh-gas; employing 
 as before, it will be observed, dotted squares to signify the 
 hypothetical character of carbon vapour. Though these syn- 
 theses have not yet been directly accomplished, they obviously 
 imply the play of elements here depicted; and we thus 
 learn that, just as in the formation of hydrochloric acid 
 gas and water-gas, respectively, univalent chlorine replaces 1 
 atom, and divalent oxygen 2 atoms, of hydrogen, just so, in the 
 production of ammonia and marsh-gas, the trivalent nitrogen- 
 atom, and the quadrivalent carbon-atom, respectively replace 
 3 and 4 atoms of the same unitary standard. 
 
 Synthesis (hypothetical) of Ammonia. 
 
 Constituent Gases. 
 
 TJieir distribution to form 
 Ammonia. 
 
 TJieir condensation 
 into Ammonia. 
 
 (_* 
 
 HH 
 
 HH 
 
 HH 
 
 H 
 
 
 N ra 
 
 II 
 
 Constituent Gases. 
 
 Or 3HH + N ra N m = 2H 3 N m . 
 
 Synthesis (hypothetical) of Marsh-gas. 
 
 Their distribution to form 
 Marsh-gas. 
 
 Tli< ir condensation 
 into Marsh-gas. 
 
 -H C IV C IV i = 
 
 j C IV ; 
 
 L J 
 
 1C j = 
 
 H 4 C" 
 
 Or 4 HH + C IV C IV - 
 
DECOMPOSITIONS ILLUSTRATIVE OF QUANT1VALENCE. 
 
 175 
 
 Nor is it only by their synthetic equations that our four typi- 
 cal compounds illustrate the dissimilar quantivalence of the 
 elementary atoms ; we shall find their equations of decomposition 
 equally instructive on this head. 
 
 We remember, for example, employing chlorine to expel oxy- 
 gen from water, nitrogen from ammonia, and carbon from marsh- 
 gas ; and, keeping in view the quantivalence of these bodies 
 respectively, we may be sure that, of the univalent chlorine, 2 
 atoms are requisite to expel from the water-gas molecule its atom 
 of divalent oxygen, 3 to remove from the ammonia molecule its 
 atom of trivalent nitrogen, and 4 to replace in the marsh-gas 
 molecule its atom of quadrivalent carbon. 
 
 Experiments, which we have already made together, have 
 established these relative values as indisputable facts ; and they 
 are compendiously displayed in the following symbolic equations. 
 
 Decomposition of Water by Chlorine. 
 
 1. Atomic expression. 
 
 nil . 77 , Hydrochloric Acid Oxygen lil 
 
 Water-qas. Its constituents. >ea " formed (without rated. 
 
 condensation}. (1 atom . 
 
 H.,0 1 
 
 Water-gas. 
 (2 molecules}. 
 
 Its constituents. 
 
 S - 
 
 Or H 2 n -f 2C1 1 - 2HC1 1 + O n . 
 2. Molecular expression. 
 
 Chlorine added. H y arocllloric Acid Oxygen lilt- 
 (2 molecules}. ( L~L.\ (1 molecule 
 
 O n O 1 
 
 Or 2H 2 U 
 
 = 4HCP + O n O n . 
 
176 
 
 FURTHER ILLUSTRATIONS OF 
 
 Ammonia 
 Gas. 
 
 Ammonia 
 
 Gas. 
 (2 molecules}. 
 
 Decomposition of Ammonia by Chlorine. 
 1. Atomic Expression. 
 
 Its con- 
 stituents. 
 
 Chlorine Hydrochloric Nitrogen 
 
 added. Acid Gas liberated. 
 
 (3 atoms}. formed. (1 atom}. 
 
 H N 111 4- Cl 1 = 
 
 Or H 3 N ra + 3C1 1 = 3HCF + N m . 
 
 2. Molecular expression. 
 
 Its con- 
 stituents. 
 
 Chlorine Hydrochloric Add Nitrogen 
 added. Gas formed. liberated. 
 
 (3 molecules}. (6 molecules}. '\ molecule}. 
 
 or 
 
 II 
 
 Or 2H 3 N ra + 3CrCr = 6HC1 1 -f N m N m . 
 
Harsh-gas. % con / 
 
 stttuents. 
 
 MOLECULAR EQUATIONS. 
 
 Decomposition of Marsh-gas by Chlorine. 
 1. Atomic expression. 
 
 Chlorine Hydrochloric Carbon 
 added. Acid Gas liberated. 
 
 (4 molecules). formed. (latom\ 
 
 177 
 
 Or H 4 C IV + 4C1 1 = 4HC1 1 + C 17 . 
 2. Molecular expression. 
 
 Marsh Gas Its con- 
 2 molecules'). stituents. 
 
 Chlorine Hydrochloric Acid Carbon 
 added. Gas formed. liberated. 
 
 molecules'). (8 molecules}. (I molecule*} 
 
 or 
 
 
 ! 
 
 Her 
 
 
 
 
 
 
 era 1 
 
 1 
 
 
 
 
 
 HCl 1 
 
 
 
 
 
 
 
 
 HCl 1 
 
 
 
 
 
 
 crci 1 
 
 
 
 
 
 
 Her 
 
 
 
 
 
 
 
 
 HCl 1 
 
 
 
 
 A 
 
 
 crci 1 
 
 
 
 
 ^ 
 
 
 
 
 
 
 HCl 1 
 
 
 
 
 
 
 
 
 HCl 1 
 
 
 
 
 
 
 crci 1 
 
 
 
 
 
 
 
 
 
 
 HCl 1 
 
 
 Or 2H 4 C IV + 4C1 I CP = 8HC1 1 
 
 i 3 
 
178 CONTRASTED QUANTIVALENCE 
 
 Each of the foregoing equations, it will be observed, is given 
 in two forms ; one atomic, the other molecular. The former is 
 the simpler, and, to the beginner, the more intelligible mode of 
 expression ; but, as it frequently depicts free gases as if com- 
 posed of isolated atoms, it is to that extent incorrect. The 
 molecular expressions represent the free gases as of their true 
 diatomic structure ; and, accordingly, though more cumbrous, 
 they are more strictly accurate symbolizations. By the in- 
 structed eye these latter will therefore be preferred; and by 
 contrasting the two forms, in the few previous cases, we obtain 
 an opportunity of familiarizing ourselves with both. 
 
 Another excellent illustration of this subject is supplied by 
 the contrasted behaviour of chlorine and oxygen, when employed 
 successively to expel iodine from hydriodic acid. Iodine, as 
 we remember, is a congener of chlorine, and, like it, univalent ; 
 whereas oxygen is our typical bivalent. Oxygen ought, there- 
 fore, to display twice the atom-expelling power of chlorine; 
 wherefore, if chlorine, as a univalent body, expels from hydri- 
 odic acid 1 atom of iodine (also univalent), oxygen ought evi- 
 dently to expel therefrom two. In other words, for 1 molecule 
 of hydriodic acid decomposed by a chlorine-atom, 2 should be 
 decomposed by an oxygen-atom. This ratio of quanti valence, or 
 atomic value in exchange, is actually established by experiment. 
 For any given weight of hydriodic acid attacked, and any given 
 number of iodine-atoms expelled therefrom, by univalent chlo- 
 rine, twice the weight and double the number are attacked and ex- 
 pelled by bivalent oxygen. For the display of these results in the 
 diagrams subjoined, we have selected the more precise molecular 
 form of expression ; which, by aid of the above models, you can 
 readily transform, if so minded, into the simpler atomic equations. 
 
 In each case we have, it will be observed, as in the previous 
 diagrams, placed first the symbol of the compound to be decom- 
 posed, next this the element added for its decomposition, and 
 afterwards, in due succession, the products, elementary and com- 
 pound, of the reaction ; interposing, when necessary, a sketch of 
 the redistribution of the atoms present, to form the new products. 
 
Hydriodic 
 
 Acid Gas. 
 
 (2 molecules]. 
 
 OF CHLOEINE AND OXYGEN. 179 
 
 Decomposition of Hydriodic Acid by Chlorine. 
 
 Hydrochloric Acid Gas j -.. 
 
 Or 2HI< + CrCl 1 = 2HC1 X -f IT. 
 
 Decomposition of Hydriodic Acid by Oxygen. 
 
 Hydriodic 
 
 Acid Gas. 
 
 (4 molecules}. 
 
 Oxygen 
 
 added. 
 
 (1 molecule}. 
 
 Their distribution to 
 
 form Water-gas and 
 
 free Iodine. 
 
 Water-gas 
 formed. 
 
 (2 molecules}. 
 
 Iodine 
 
 liberated. 
 
 (2 molecules 
 
 HI 1 
 
 
 
 
 HI 1 
 
 4- 0"0 n 
 
 
 HP 
 
 
 
 HP 
 
 H 
 
 
 r 
 
 
 
 0" 
 
 
 
 H 
 
 r 
 
 
 
 
 H 
 
 r 
 
 O n 
 
 
 
 H 
 
 
 i 1 
 
 Or 4HI 1 -f O n O n = 2H 2 O n + 2I 1 ! 1 . 
 
 Once more, if we compare the quantivalent powers of univalent 
 hydrogen and bivalent oxygen, as manifested in the quantity of 
 trivalent nitrogen which they respectively engage in forming 
 ammonia and nitrous acid, we obtain another exact and beautiful 
 confirmation of the law of quantivalence. This is displayed in 
 the following diagram, in which, it will be observed, the molecule 
 of nitrous acid gas is represented in dotted outline, our usual 
 sign of doubt ; the vapour-density of this body (like that of nitric 
 acid) remaining still to be determined : 
 
180 QUANTIVALENTIAL EQUILIBRIUM OF THE XITROXTGEN SERIES. 
 
 Comparison of the Structure of Ammonia and Nitrous Acid. 
 
 Ammonia Its Constituent Nitrous Acid Its Constituent 
 
 (2 molecules). Atoms. (1 molecule). Atoms. 
 
 
 
 
 H 
 
 
 
 
 H 3 N m 
 
 H 
 
 
 N<* 
 
 
 
 
 
 H 
 
 
 H 
 
 
 TT "M"ni 
 
 H 
 
 
 N- 
 
 
 
 
 
 H 
 
 This diagram clearly sets before us, on one side (in ammonia), 
 2 atoms of trivalent nitrogen reciprocally fixing, and fixed by, 6 
 atoms of univalent hydrogen ; while to the right, 2 atoms of tri- 
 valent nitrogen, in nitrous acid, fix, and are fixed by, 3 atoms only 
 of bivalent oxygen. Multiplying, in each case, the numbers of 
 atoms engaged, by their coefficients of quantivalence, we have 
 on both sides alike, 6 units of quantivalence balancing 6. 
 
 Thus for ammonia we have N m x 2 = VI balancing H r x 6 = VI , 
 
 While for nitrous acid we have N m x 2 = VI balancing O n x 3 = VI . 
 
 It is by no means, however, to be supposed that the atom- 
 fixing or atom-expelling power of the elementary bodies, or, as we 
 have agreed to term it, their quantivalence, is under all circum- 
 stances, and in all the compounds which they form, invariably 
 exerted to its fullest extent. Were the chemical combination of 
 elementary atoms governed in this respect by a fixed and un- 
 bending law, it would evidently be impossible for any two given 
 elements to unite in any other proportion than that implying 
 their complete reciprocal satisfaction. Trivalent nitrogen and 
 
DIAGRAMMATIC REPRESENTATION THEREOF. 
 
 181 
 
 divalent oxygen, for example, could, under such a law, unite 
 only in the proportion of two atoms of the former body to three 
 of the latter to form the compound N0", which we have just 
 been comparing with ammonia ; whereas we already know that 
 these two elements form, by their varied combinations, no less 
 than 5 different compounds, which we have studied under the 
 name of the Nitroxygen series (compare p. 183). Of this series one 
 member only, viz., N0", enjoys what may be termed a perfect 
 quantivalential equilibrium ; that is to say, a complete reciprocal 
 satisfaction of their quantivalence by the elements present. 
 
 It has been recently pointed out (comp. Preface, p. vi.) 
 that, while this perfectly balanced, or self-satisfied body, N0" , 
 is the central body, or pivot, of the nitroxygen series, this series 
 itself, taken collectively, presents an equally perfect equipoise, 
 or quantivalential symmetry ; its four remaining members being 
 disposed in pairs, one pair on either side of the centre, the 
 members of the left-hand pair containing an excess of N m quanti- 
 valence unbalanced by O u ; while those of the right-hand pair, 
 on the contrary, possess an excess of O n quantivalence uncom- 
 pensated by N m . 
 
 The interest of this observation (now, I believe, first published) 
 lies mainly in the curious fact, that these opposite excesses exactly 
 match, or, so to speak, inversely reflect each other ; a fact which 
 its observer has constructed the following diagram to display. 
 
 Nitroxygen Compounds. 
 Tlieir Quantivalential Composition. 
 
 Hypo-oxygenic. 
 
 Equiquantic. 
 
 Hypo-nitrogenic. 
 
 Nitrous Oxide. Nitric Oxide. Nitrous Acid. Hyponitric Acid. Nitric Acid. 
 
 Their Q.uantiquicalential Proportions. 
 
 6 to 2 3 to 2 6 to 6 3 to 4 
 
 Their Quantivalential Deficiencies. 
 
 4101 
 
 6 to 10 
 
182 EQUIQUANTICITY NUMERICAL POTENTIAL. 
 
 The foregoing diagram, it will be observed, is so arranged as 
 to exhibit prominently, in the centre of the series, that one of the 
 five nitroxygen compounds in which the opposite quantivalential 
 or atom-fixing powers are equally balanced ; a condition which, 
 to avoid periphrasis, we will agree, if you please, to term equi- 
 quantic. From this central compound the others are shown 
 falling away on either side, to represent their progressive declen- 
 sion from equiquanticity. It will be observed that each case or 
 decline (case is from the Latin casus, a fall) that takes place on the 
 side of deficiency in N m quantivalence, is matched by a corre- 
 sponding case, or decline, taking place on the side of deficiency in 
 0" quantivalence ; so that the series is perfectly symmetrical. 
 
 It will be also observed that the deficiency of quantivalence 
 in each of the terminal compounds of the series (nitrous oxide 
 and nitric acid) is = 4 ; which deficiency diminishes to = 1 
 only, for each of the two compounds (nitric oxide and hyponi- 
 tric acid) that intervene between the terminals and the central 
 or equiquantic compound (nitrous acid). 
 
 There is certainly much beauty in the perfect symmetry and 
 absolute equipoise of this natural chemical group ; with its two 
 equal wings spreading, so to speak, from the central body, each, 
 in respect of its quantivalential relations, the very counterpart 
 of the other. 
 
 Similar deficiencies of quantivalence occur, of necessity, in 
 every other series of binary compounds exemplifying, like the 
 nitroxygen series, the law of multiple proportions. Such de- 
 viations are also frequently observed in compounds of higher 
 complexity, belonging to series of greater extent. The condi- 
 tions of their occurrence, and the characters they impress upon 
 the compounds thus unequally composed, are subjects for expe- 
 rimental study, and philosophical meditation, of the very deepest 
 interest. We shall hereafter find that bodies which are of mir 
 mericalty equiquantic structure are not always, on that account, 
 the most stable ; while, on the other hand, numerical deficiencies 
 of quantivalence often coexist with what may be termed potential 
 equiquanticity. For example, in the nitroxygen series itself, 
 the central equiquantic body, N0, is not, as might be ex- 
 
DISTINCTION OF CHEMISM FROM QUANTIVALENCE. 183 
 
 pected, the most stable compound of the series ; on the contrary, 
 it breaks up readily under various influences; whereas the 
 terminal member of the left wing, N0 n , notwithstanding its 
 hypo-oxygenic structure, possesses remarkable stability. Again, 
 N0, the terminal member of the opposite or hypo-nitrogenic 
 wing, though it is numerically on a par, in point of quantiva- 
 lential deficiency, with N0 U (the deficit being represented by 
 4 in both cases), instead of possessing the stability of that body, 
 cannot be vaporized without breaking up into compounds of a 
 lower order. These apparently anomalous facts depend upon 
 conditions into the consideration of which we cannot here 
 enter. At present it must suffice us to remark, in passing, 
 that, cceteris paribus, compounds in which the quantivalents are 
 numerically balanced are more readily formed, and of more 
 stable character, than those in which either constituent is in 
 quantivalential excess. 
 
 At this point a few words may be usefully interposed to 
 guard the student from misconceiving the true import of the 
 term quantimlence. This word is employed to designate the 
 particular numerical atom-compensating power inherent in each 
 of the elements, and this power of theirs must by no means be 
 confounded with the specific intensity of their respective chemical 
 activities. Thus, for example, nitrogen, phosphorus, and arsenic 
 are all of them In'valent bodies ; but this equality of their atom- 
 compensating values does not imply that they are all endowed 
 with equal avidity for this or that element say oxygen or hydro- 
 gen, for example. Because one ^wa^n'valent carbon atom can fix 
 four atoms of hydrogen, of which one only is fixed by umvalent 
 chlorine, we are not, therefore, to suppose that the attraction 
 of carbon for hydrogen is necessarily more intense than that of 
 chlorine, in the ratio ^. On the contrary, in this case, the ele- 
 ment which seizes the lesser proportion of hydrogen, seizes it 
 with by far the greater degree of energy. For another and 
 equally cogent illustration of this truth we may refer to our very 
 first experiment. In this, as you remember, we saw potassium 
 (a univalent body, as we shall presently learn), take oxygen from 
 its combination with hydrogen (also univalent) in water ; equalit 
 
184 DISTINCTION OP OHEMISM FROM MOLECULE-FORMING POWER. 
 
 of quantivalence coexisting, in this case, with a strongly-marked 
 difference in the degree or intensity of that peculiar endow- 
 ment which is commonly denominated chemical attraction, 
 but to which the shorter and more abstract appellation 
 Chemism may be more safely and appropriately assigned. The 
 term chemism is derived from the same root which also 
 supplies chemistry itself with its name (comp. p. 2) ; and 
 chemism seems a fitting designation for that peculiar power, 
 or endowment, which constitutes the special subject of the 
 science analogously named. The phenomena of chemism, and 
 the conditions of its varying intensity, have hitherto but cur- 
 sorily engaged our attention, and rather as incidental to expe- 
 riments made with other subjects of study in view, than as 
 themselves claiming immediate notice. We shall have to examine 
 those phenomena in their turn hereafter; and we shall then 
 learn, amongst other things, how much the relative chemism of 
 bodies varies, so far at least as its results are concerned, with 
 the varying physical conditions under which it is called into 
 play. We merely refer to this subject here, that we may note 
 in passing, the strong lines of demarcation which divide the 
 chemism of the elements from their molecule-forming power, on 
 the one hand, and, on the other, from their atomic quantivalence. 
 With reference to this latter property of the elementary bodies, 
 and to the relations which subsist between its laws and those 
 of combination in multiple proportions, much as these subjects 
 tempt us to further investigation, this must, for the present, 
 stand adjourned. We must, indeed, become acquainted with 
 a greater variety of chemical compounds before we can further 
 pursue with advantage this branch of our inquiry. The scanty 
 materials as yet at our disposal do not enable us fully to 
 appreciate the assistance afforded to modern chemical research 
 by the development of the laws which govern the unequal 
 molecule-forming and atom-fixing powers of the elementary 
 bodies. In these respects modern chemistry itself is, indeed, 
 in a state of rapid transition ; each year not to say every day 
 adding something to our knowledge of the atomic and mole- 
 cular endowments of the elementary bodies and their com- 
 
TABLE OF ATOMIC AND MOLECULAR WEIGHTS. 
 
 185 
 
 pounds ; of the analogies which tend to enrich with new 
 members the groups they respectively represent ; and of the 
 exceptions apparent or real which, for the present, obscure 
 the boundaries of those groups. For the perfect elucidation of 
 these deeply-interesting problems we must await the "long 
 results of time " and must all be contented, whether teaching 
 or taught, humbly to labour and learn, side by side, in the school 
 of the great Teacher NATUKE. 
 
 The following Table is an extension of the one above given, 
 comprising the four typical elements only. 
 
 TABLE of the MINIMUM MOLECULE-FORMING and MOLECULAR 
 WEIGHTS of SEVERAL ELEMENTS, with their coefficients of 
 quanti valence . 
 
 Names of the 
 typical elements 
 and their conge- 
 ners ; preceded by 
 that of the stand- 
 ard element. 
 
 Minimum molecule- 
 forming weights; 
 also called 
 combining numbers, 
 and atom-weights ; 
 with their 
 atomic symbols and 
 coefficients of quanti- 
 valence. 
 
 Molecular weights and 
 their symbols ; showing the atomic 
 structure of the elementary 
 molecules in their uncombined 
 state. 
 
 1. 
 
 2. 
 
 3. 
 
 Hydrogen 
 
 H = 1 
 
 HH =2 
 
 Chlorine . . 
 Bromine . 
 Iodine . . 
 
 or = 35-5 
 
 Br 1 = 80 
 I 1 = 127 
 
 Cl'Cl 1 - 71 
 Br'Br" = 160 
 IT = 254 
 
 Oxygen . . 
 Sulphur . . 
 Selenium 
 
 O n = 16 
 S n = 32 
 
 Se n = 79 
 
 O n O n = 32 
 
 S n S n = 64 
 Se n Se n - 158 
 
 Nitrogen . 
 Phosphorus . 
 Arsenic . . 
 
 N m - 14 
 P ra = 31 
 
 As m = 75 
 
 N m N ra = 28 
 
 pmpmpmpin _ -^4 
 
 As ra As m As m As in = 300 
 
 Carbon . . 
 Silicon . . 
 Titanium . . 
 
 C IV - 12 
 Si w - 28-5 
 Ti IY = 50 
 
 QIVQIV ? =24 
 
 Si^Si" ? =57 
 Ti iv Ti iv 9 =100 
 
186 EXPLANATION OF THE FOREGOING TABLE. 
 
 In this Table all the simple bodies we have had under our 
 notice are set forth in the first column, with their minimum 
 molecule-forming weights, or atom-weights, set against them in 
 the second column ; while, in the third column, these atomic 
 expressions and weights are converted by appropriate modifi- 
 cations into molecular symbols. The coefficients of quantiva- 
 lence are given in both columns ; and hydrogen, as before, is 
 placed alone at the head of the Table, as the unitary standard to 
 which all the other bodies are referred. These latter, it will 
 be observed, are disposed in the four typical groups, now (I 
 trust) so familiar to us all. 
 
187 
 
 LECTURE XI. 
 
 Principles established for 13 elements, how far applicable to the remaining 
 48 method of determining the atom-weights of elements, neither volatile 
 themselves, nor capable of forming volatile compounds with hydrogen 
 value of volatile chlorine-compounds for this purpose also of volatile oxy- 
 gen and nitrogen compounds when such exist illustrations of the prin- 
 ciple supplied by such as are volatile of the chlorine, oxygen, and nitrogen 
 compounds of hydrogen, sodium, and potassium atom-weights and quanti- 
 valence of sodium and potassium as determined by the balance prospect 
 of volumetric corroboration of these results further illustrations mercury 
 bismuth tin- their vapour-densities pondend composition of their 
 respective chlorides deductions therefrom of their respective atom- 
 weights and coefficients of quanti valence universal applicability of this 
 method its liability to error in one respect further data requisite to 
 obviate this such data how to be obtained curious exceptional relation 
 of the atomic and molecular weights of mercury as also of cadmium 
 contrast presented by these metals in this respect to phosphorus and 
 arsenic contrasted symbolization of these bodies monatomic, diatomic, 
 tetratomic, and polyatomic molecules corresponding anomalies in the 
 condensing powers of the respective atoms of these bodies physical aids 
 to the determination of atom-weights specific-heat method isomorphic 
 method cases in which they are of special value table of the elements, 
 with their atomic and molecular symbols, weights, volumes (so far as 
 known) and coefficients of quantivalence one exception to the univer- 
 sality of the symbols appeal, on this subject, to scientific France. 
 
 THUS far, in our study of the relations of the elements, as to 
 their combining proportions, volumetric arid ponderal, and as to 
 their chemical powers, molecule-forming and atom-fixing, we 
 have confined our attention to a few typical bodies and their 
 immediate analogues. Hydrogen has stood apart as our unitary 
 standard ; while chlorine, oxygen, nitrogen, and carbon, each 
 grouped with two congeners thirteen bodies in all have sup- 
 plied us with the whole of the facts we have employed, to form 
 the basis of our demonstration, and to establish that symbolic 
 language which, expanding as we proceeded, has served both 
 to facilitate our investigations, and tersely to embody our 
 results. 
 
 We have now to step beyond this limited sphere of inquiry, 
 
188 EXTENSION OF THE INQUIRY TO OTHER ELEMENTS. 
 
 and to determine how far our methods of research, our principles 
 of classification, and our symbolic system of record, derived 
 from the study of thirteen elements only, are applicable to the 
 investigation and co-ordination of the sixty-one elements which 
 constitute our planet. 
 
 For such of the typical elements and their congeners as are 
 permanently gaseous, or readily vaporizable, we have had no 
 difficulty in directly fixing the relative gas-volume weights or 
 vapour-densities, by weighing one litre of each gas, or vapour, 
 in comparison with the corresponding measure of hydrogen 
 taken as unity. These relative gas-volume-weights or specific 
 gravities of the volatile elements, we have found generally to 
 represent their minimum molecule-forming weights ; also 
 designated their atom- weights, or combining numbers. 
 
 But the exceptional cases of phosphorus and arsenic, whose 
 gas-volume-weights or specific gravities we found to be double 
 their molecule-forming or atom weights, obliged us to regard 
 the specific gravity of gaseous bodies as but an uncertain criterion 
 of their combining numbers ; though, on the other hand, the ex- 
 treme simplicity of the ratio (2 to 1) subsisting between the 
 specific gravity and atom-weight of phosphorus and arsenic 
 respectively, afforded us a good prospect, if not of certainly 
 making out, by the specific gravity method, the true combining 
 atom-weights of volatile elements, at all events of arriving at 
 figures standing in very simple ratios thereto, and therefore 
 likely, if taken in conjunction with other indications, to assist 
 in guiding us to the desired result. 
 
 The method upon which, in the case of phosphorus and 
 arsenic, we were compelled to fall back, for the determination 
 of their true atom-weights, consisted in ascertaining, by means 
 of the balance, the weight of each of those bodies, respectively, 
 contained in the normal product- volume (2 litres) of the com- 
 pound gas formed by it with hydrogen. 
 
 In the case of carbon we resorted to the same method for a 
 different reason, viz., the impossibility of volatilizing carbon, so 
 as to ascertain directly its relative gas-volume weight. We 
 therefore determined, by means of the balance, the proportion 
 
ATOM-WEIGHT DETERMINATIONS OF NON-VOLATILE ELEMENTS. 189 
 
 by weight of carbon contained in two litres of that one of 
 its gaseous compounds with hydrogen in which it is least 
 abundant ; and this weight we took, on the ground of analogy, 
 as the true molecule-forming minim, or combining weight, 
 otherwise called atom-weight, of carbon. 
 
 So far as the other non-volatile elements are obtainable in 
 gaseous combination with hydrogen, the method which thus 
 served us, in the case of carbon, is evidently applicable, in their 
 cases also, to determine their respective molecule-forming 
 minims, or atom-weights. 
 
 Unfortunately, however, a large majority of the elementary 
 bodies, while resembling carbon as to their non-volatility under 
 any treatment in our power to employ, differ from carbon in the 
 circumstance that they have hitherto resisted our endeavours 
 to bring them into combination with hydrogen. 
 
 The problem now before us is, therefore, how to determine 
 the molecule-forming minims, or atom-weights, of those numer- 
 ous bodies which neither yield vapours to be directly weighed, nor 
 form gaseous or volatile compounds with our unitary standard, 
 hydrogen ? 
 
 It is a well-known mathematical axiom, that "things which 
 are equal to the same thing are equal to one another :" and this 
 axiom is as true with respect to chemical as it is with respect 
 to mathematical quantities. 
 
 A little consideration will satisfy us that this self-evident 
 principle, coupled with the knowledge which we now possess, 
 of molecular and atomic values, greatly enlarges the otherwise 
 restricted range of facts, available as sources of information 
 concerning the atom-weights of the elements still remaining to 
 be investigated. 
 
 Thus, knowing as we do the molecule-forming minim-weight, 
 and the atom-fixing or quantivalent power, of chlorine, relatively 
 to hydrogen taken as unity ; and knowing also that the product- 
 volume of volatile chlorine-compounds corresponds precisely 
 with that of volatile hydrogen-compounds, such volume being, for 
 both classes of compounds, dilitral ; it is clear that, for elements 
 which do not enter into combination with hydrogen, but which 
 
190 USE OF CHLORINE-, OXYGEN-, AND NITROGEN-COMPOUNDS 
 
 do form volatile or volatilizable compounds with chlorine, \ve 
 may fall back on these chlorine-compounds, and, from the pon- 
 deral analysis thereof, deduce the molecule-forming minims or 
 atom-weights sought, as readily and certainly as if the missing 
 hydrogen-compounds themselves were at our disposal. 
 
 The axiom which we have quoted as placing chlorine on a par 
 with hydrogen for the purpose in view, is, of necessity, equally 
 applicable (mutatis mutandis) to show oxygen and nitrogen avail- 
 able to the same end. 
 
 Of course, in so using oxygen and nitrogen, we must take 
 duly into account their respective molecule-forming minim- 
 weights and atomic quantivalence, as to both of which values, 
 these gases (we know) differ, both from each other and from 
 chlorine. But, these being allowed for, the volatile oxygen and 
 nitrogen compounds of the non-volatile elements are as avail- 
 able as their corresponding chlorine-compounds, nay, even as 
 their hydrogen-compounds themselves (when such exist), for 
 the determination of their molecule-forming minims or atom- 
 weight s. 
 
 Among the experiments which we have already made there 
 happens fortunately to be a series capable of furnishing appro- 
 priate and satisfactory illustrations of our present theme. 
 
 We employed, as you well remember, the alkali-metals, 
 sodium (Na) and potassium (K), to withdraw chlorine from 
 hydrochloric acid (HC1 1 ), oxygen from water (H 2 U ), and nitro- 
 gen from ammonia (H 3 N m ). 
 
 The compounds of Na and K which we thus formed respec- 
 tively with chlorine, with oxygen, and with nitrogen, we were 
 led, by analysis, to represent in formulae, which displayed most 
 clearly their exact structural analogies with the three hydro- 
 gen compounds hydrochloric acid, water, and ammonia from 
 which they respectively originated. 
 
 In the following table we have placed in the first column the 
 names, symbols, coefficients, and atom-weights of chlorine, 
 oxygen, and nitrogen ; in the second column, the formulae of 
 their respective combinations with hydrogen ; and in the third 
 and fourth columns respectively, the formulae of their analogous 
 
FOR THE DETERMINATION OF ATOM-WEIGHTS. 
 
 191 
 
 combinations with sodium and potassium. Thus, reading the 
 table horizontally, we have, in the first row three chlorides, in 
 the second row three oxides, and in the third row three nitrides, 
 brought into direct comparison. 
 
 COMPOUNDS OF CHLORINE, OXYGEN, AND NITROGEN, WITH 
 HYDROGEN, SODIDM, AND POTASSIUM. 
 
 Names, symbols, Names, symbols, coefficients of quantivalence, and 
 
 coefficients of quan'ti- I combining weights of the elements taking part, each in 
 
 valence, and combining | three of the Compounds formed ; with 
 
 weights of the three ' formulae of the products. 
 
 elements existing in all 
 the Compounds 
 formed. 
 
 Hvdrogen H = l. 
 
 Sodium Na = 23. 
 
 Potassium K 
 
 = 89. 
 
 Chlorine = CP, 35'5 
 
 HC1 1 
 
 NaCP 
 
 KCP , 
 
 
 Oxygen = O u 16. 
 
 HgO" 
 
 Na 2 O" 
 
 K 2 n 
 
 
 Nitrogen = N m 14. 
 
 H 3 N m 
 
 Na 3 N ra 
 
 K 3 N ra 
 
 
 Now, H representing unity as a matter of course, why have we 
 in this table made Na = 23 and K = 39 respectively ? Because 
 these are the respective weights of those two bodies which, in 
 each of the compounds figured in the table, exactly replace, and 
 do the work of, 1 part by weight of hydrogen. 
 
 Carrying the eye along the chlorine line, we see that one part 
 by weight of hydrogen in HC1 1 is replaced, and its atom-fixing 
 duty fulfilled, by 23 parts by weight of sodium in NaCl 1 , and by 
 39 parts by weight of potassium in KC1 1 ; or, to express the same 
 thing more succinctly, we see Cl 1 = 35 -5 satisfied, equally and 
 indifferently, by H = 1, by Na = 23, and by K = 39. 
 
 Again, carrying the eye along the oxygen line, we see H 2 = 2 
 replaced, and its atom-fixing function performed by Na^, or 
 23 x 2 - 46 ; and by K 2 , or 39 x 2 = 78. 
 
 Lastly, glancing similarly along the nitrogen line of the table, 
 we see H 3 = 3 replaced by Na 3 , or23x3 = 69; and by K 3 , or 
 39 x 3 = 117. 
 
 Short of a demonstration by actual experiment, it is difficult 
 to conceive stronger grounds than those afforded by these 
 
192 DETERMINATION OF ATOM-WEIGHTS OF SODIUM AND POTASSIUM. 
 
 comparisons, for regarding Na and K as univalent bodies, = 
 Na 1 and K 1 , and assigning to them respectively 23 and 39, as 
 their atomic weights. 
 
 Happily, as both Na 1 and K 1 are volatile metals, and as their 
 chlorides and oxides are also volatile (at furnace temperatures), 
 we may hope to obtain, in due time, direct experimental proof 
 that their respective atom-weights are 23 and 39, and that their 
 atoms are univalent. 
 
 Such proof would be evidently obtained, should we succeed in 
 determining the weights of the product- volumes of Na 1 CP and 
 K'CF, and should we find these weights equal to 23 -f- 35*5 = 58-5 
 for Na'Cl 1 , and to 39 4- 35-5 = 74 5 for K'Cl'. These proofs 
 would, of course, be powerfully corroborated should we further 
 succeed in weighing the dilitral volumes of Na/,0 11 and K.^O", and 
 should these weights prove, conformably with our anticipations, 
 to be 23 x 2 + 16 = 62 for the former oxide, and 39 x 2 -f 16 
 = 94 for the latter. 
 
 It is hardly necessary to add, that assurance would be made 
 doubly sure, should we at any future time succeed in obtaining 
 volatile hydrogen-compounds of sodium and potassium, and 
 should these prove, on analysis, to have, respectively, the compo- 
 sition and weight, HNa 1 -1 + 23-24, and HK 1 - 1 + 39 - 40, 
 in one dilitral product- volume (comp. p. 199). 
 
 In the present eminently transitional state of chemistry we do 
 not possess any such absolutely irrefragable proofs of the atom- 
 weights of Na 1 and K 1 ; and, pending the researches which, in a 
 hundred laboratories, are constantly in progress for the more 
 strict solution of these and many cognate problems, we provi- 
 sionally adopt, not for Na 1 and K 1 only, but for numerous other 
 elements, atom-weights resting on strong grounds of probability, 
 but still awaiting their final and unconditional verification. 
 
 For bodies which appear likely to resist, in the future as in the 
 past, all attempts to obtain and weigh the vapour, either of them- 
 selves or of any of their compounds, we must of course expect to 
 rely permanently, for the determination of their respective atom- 
 weights, upon the ponderal analysis of their fixed combinations 
 with one or more of the typical elements (as, for instance, 
 
UNCEKTAINTT OF PONDEKAL ATOM-WEIGHT DETERMINATIONS. 193 
 
 vvith chlorine and oxygen) ; the minim-weight which combines 
 with 35' 5 of univalent chlorine, and half the minim-weight 
 which combines with 16 of divalent oxygen, being adopted, in 
 each such case, as the atom-weight sought. Atom-weights thus 
 determined exclusively by aid of the balance must, of course, 
 remain permanently liable to the same uncertainty as pro- 
 visionally attends like purely ponderal determinations of the 
 atom-weights of bodies which, like Na 1 and K 1 , we hope here- 
 after to submit to volumetric verification. 
 
 A few examples, chosen from among the elements to which our 
 attention has not yet been turned, will assist us in firmly grasping 
 the method of fixing, both by ponderal. and volumetric determi- 
 nations, the respective molecule-forming minims, or atom-weights 
 of those elements, and their atomic quantivalence. 
 
 For this purpose we will select three familiar metals : Mer- 
 cury, with which our pneumatic trough is filled, and which we 
 see in every looking-glass and every barometer; Bismuth, an 
 ingredient of common pewter and of the well-known fusible 
 alloy ; Tin, with which our kitchen implements are coated. 
 
 Neither of these metals has yet been obtained in combination 
 with hydrogen ; but they all of them combine readily with chlo- 
 rine ; and, fortunately, their compounds with chlorine are volatile, 
 so that we are enabled to determine their vapour-densities. 
 
 These have been accurately ascertained ; as also have the 
 proportions by weight in which their respective constituents are 
 combined therein. These facts are set forth in the following 
 table: 
 
 Gas-volume- Proportions of the Con- 
 
 weight, stituents therein 
 
 Metal. Chlorine. 
 
 Chloride of mercury .... 135-50 .... 100 .... 35-5 
 
 bismuth .... 157-25 .... 104 .... 53-25 
 
 tin 130- .. .. 59 .. .. 71- 
 
 From this table it is of course easy to compute the weight 
 and composition of the normal product-volume of each of these 
 compounds. For, this volume being, as we know, dilitral, while 
 
 K 
 
194 FURTHER EXAMPLES MERCURY BISMUTH TIN. 
 
 the unit-volume referred to in the above table is monolitral, we 
 have only to multiply the above figures by 2 to obtain the corre- 
 sponding dilitral, or, as we now say, the molecular expressions. 
 
 These calculations have been made, and they give us the fol- 
 lowing results. 
 
 The dilitral or product-volume of chloride of mercury is found 
 to contain 2 unit-volumes, representing (as we now know) 2 
 atoms, of chlorine ; or, by weight 35'5 x 2 = 71 parts of chlorine ; 
 from which fact we at once learn that the atom of mercury is 
 bivalent. With these 71 parts of chlorine are associated, in the 
 product-volume of mercurial chloride, 200 parts by weight of 
 mercury ; which fact acquaints us with the atom- weight of 
 mercury, whose Latin name is hydrargyrum. To this metal we 
 accordingly assign the symbol Hg 11 = 200. 
 
 Again, the dilitral or product-volume-weight of chloride of 
 bismuth contains 3 unit-volumes, representing 3 atoms, of chlo- 
 rine ; or, by weight, 35'5x3 = 106'5 of chlorine ; whence we learn 
 that bismuth is trivalent. With these 106*5 parts of chlorine 
 are associated, in the product-volume of the bismuth-chloride, 
 208 parts by weight of bismuth ; bespeaking, for the bismuth- 
 atom, the symbol and weight Bi m = 208. 
 
 Lastly, in the dilitral or product- volume of tin chloride, analy- 
 sis reveals 4 unit-volumes = 4 atoms of chlorine ; or; by weight, 
 35*5 x 4 = 142 of chlorine; proving tin to be quadrivalent, like car- 
 bon. With these 142 parts of chlorine we find associated in the 
 product-volume of the tin-chloride, 118 parts of tin, in Latin 
 stannum ; whose atom we accordingly symbolize as Sn = 118. 
 
 This method is of universal applicability ; indeed, of all the 
 methods available for use it is the one most free from liability 
 to error. It implies, however, the assumption that the product- 
 volume in each case contains one atom only, not two or more 
 atoms, of the metal under examination. Further data are, therefore, 
 necessary in these cases, in order that this doubt may be set at rest. 
 And such data are fortunately supplied by methods to which we 
 shall have occasion to refer in the sequel. (Comp. pp. 199 200.) 
 
 We have already seen, in the cases of phosphorus and arsenic, 
 that the relative volume-weights of the volatile elements, though 
 
EXCEPTIONAL VAPOUR-DENSITY OF MERCURY AND CADMIUM. 195 
 
 they usually coincide with the atom- weights, are liable to deviate 
 therefrom exceptionally ; so that they cannot be relied on as 
 certain indices of the molecule-forming minims. 
 
 Further proof of this fact is afforded by one of the three 
 metals just examined, viz., mercury the only one of the three, 
 as it happens, whose vapour-density has been determined. The 
 vapour-density of mercury does not coincide with its atom- 
 weight or molecule-forming minim. Like phosphorus and ar- 
 senic, mercury is in this respect an exception to the usual rule. 
 Curiously enough, however, the deviation from type in the case 
 of mercury (which the metal Cadmium in this respect resembles) 
 is in precisely the opposite direction from that in which phos- 
 phorus and arsenic diverge therefrom. 
 
 While the atom-weights of phosphorus (P m ) and arsenic (As) 
 express respectively only half the vapour-densities of these 
 elements, the atom-weights of mercury (Hg n ) and cadmium (Cd u ) 
 represent, on the contrary, double the vapour-densities of those 
 bodies respectively. As, therefore, for the correct symbolization 
 of the P m and As atoms, we had to halve our usual monolitral 
 squares, so, to represent faithfully the atoms of Hg 11 and Cd n , we 
 have now to employ a double square our ordinary dilitral symbol. 
 
 For the clearer comprehension of these facts they are set forth 
 collectively in the following diagrammatic and descriptive table : 
 
 SYMBOLIZATION of CERTAIN ELEMENTS, STANDARD & EXCEPTIONAL. 
 
 1. Standard. 
 Hydrogen = unity. 
 
 Characters : 
 
 (a.) Atom- weight equal to specific gravity. 
 (&.) Molecular structure diatomic. 
 
 Symbols : 
 
 Atomic Molecular 
 
 (monolitral,. (diatomic and dilitral}. 
 
 Y. 2 
 
196 EXCEPTIONAL ATOMIC AND MOLECULAB SYMBOLIZAT10N 
 
 2. Exceptional, 
 a. Phosphorus and Arsenic. 
 
 Characters : 
 
 (a.) Atom- weight equal to half specific gravity. 
 (6.) Molecular structure tetratomic. 
 
 Symbols : 
 
 Atomic 
 (liemilitral}. 
 
 Molecular 
 (tetratomic and dilitral]. 
 
 y8. Mercury and Cadmium. 
 
 Characters : 
 
 (a.) Atom-weight equal to twice specific gravity. 
 (&.) Molecular structure monatomic. 
 
 Symbols : 
 
 Atomic 
 (dilitral}. 
 
 Molecular 
 (monatomic and dilitral}. 
 
 The most remarkable feature in this table is the coincidence it 
 brings out between the atom and the molecule of the respective 
 metals, mercury and cadmium ; whose molecular structure is thus 
 represented as monatomic. On this point, however, we must be 
 on our guard against attributing to the results of a particular 
 system of symbolization a deeper significance than they really 
 possess. It is only in our hypothetical conception, not as 
 a result of experimental determination, that the molecule of 
 hydrogen is regarded as diatomic. All that we really know of 
 
OF PHOSPHORUS AND ARSENIC, MERCURY, AND CADMIUM. 197 
 
 the structure of hydrogen is, that its free molecule, HH, represents 
 double the value of its combining particle, H, in point of weight, 
 magnitude, and numerical complexity of structure. We assume 
 H to be an ultimate atom merely for simplicity's sake. The 
 smallest quantity in which hydrogen combines may, for aught 
 we know to the contrary, be a cluster numbering ten, or a hun- 
 dred, or a thousand, or a million, of really ultimate atoms ; and 
 consequently, the molecular symbol HH merely implies that, 
 whatever number of ultimate atoms there may be in H, from 
 1 upwards, the free hydrogen molecule contains twice that num- 
 ber. Or, putting it in algebraic form, if n represent the nume- 
 rical constitution of the molecule HH, then the similar constitution 
 
 of the (so called) atom, H, is represented by Similarly, the 
 molecular structure of mercury, Hg, and of cadmium, Cd, relatively 
 
 to that of HH = n, is for mercury Hg = _!L and for cadmium 
 
 2 
 
 Cd = ; so that it is but the assumption that the hydrogen 
 
 molecule is diatomic, which obliges us to describe the mercury 
 and cadmium molecules as mowatomic. These molecules may 
 with equal probability be assumed to be of million-fold com- 
 plexity, if only care be taken to raise in a proportionate degree 
 our conception of the respective complexities of the so-called 
 diatomic and tetratomic molecules. 
 
 We need hardly point out that, in the exceptional cases of 
 mercury and cadmium, the volume-condensing power of the 
 atoms deviates from its ordinary relation to the quantivalent 
 power ; and that this deviation is in the sense opposite to that 
 which we have already pointed out in the cases of phosphorus 
 and arsenic. As the atoms of trivalent phosphorus and arsenic 
 are of only half the normal volume, the product-volumes of their 
 respective compounds are (so to speak) less crowded, by exactly 
 half a volume, than the product-volumes of the corresponding com- 
 pounds of their prototype, trivalent nitrogen. By parity of 
 reasoning, as the atoms of bivalent mercury and cadmium 
 
198 PHYSICAL METHODS OF INVESTIGATING ATOM-WEIGHTS. 
 
 are of double the normal volume, the product-volumes of their 
 respective compounds are packed (so to speak) with one volume 
 more than the product- volumes of the corresponding compounds 
 of their prototype, bivalent oxygen. 
 
 These anomalies are, no doubt, troublesome, as impairing the 
 unity of our symbolic language, and burdening the memory 
 with several exceptional details. With the foregoing explana- 
 tions, however, they cannot lead us to form erroneous concep- 
 tions and this is the principal evil to be avoided. 
 
 As for their philosophical interpretation, this belongs to the 
 future. They may turn out to be typical facts, round which 
 many others of like kind may come hereafter to be grouped ; 
 and they may prove to be allied with special properties, or de- 
 pendent on particular conditions, as yet unsuspected. 
 
 We shall heartily welcome, as we must patiently await, such 
 explanations hereon as time may have in store ; merely making 
 it oui % care, meanwhile, faithfully to record the facts as they 
 stand. 
 
 From the foregoing remarks it appears that, even for elements 
 which are volatile, and of which the gas volume-weights (or 
 specific gravities) can be immediately determined, this direct 
 method of ascertaining their respective molecule-forming 
 minims, or atom-weights, is subject to error, on account of their 
 liability to such anomalies as we have met with in the cases just 
 cited. 
 
 It is by first obtaining the gaseous or volatile compound formed 
 by an element, either with hydrogen itself, or, failing this, with 
 some element bearing thereto known weight and volume ratios, 
 and by then ascertaining its vapour-density and composition, so as 
 to learn how much of the element under investigation exists in 
 the dilitral volume of such compound, that the atom-weights of 
 the elements, whether volatile or fixed, can be most safely and 
 certainly determined. 
 
 To this, the purely chemical method, certain physical modes 
 of research afford, in doubtful cases, very acceptable aid. 
 
 Much light, for example, is thrown on the atom -weights of the 
 
SPECIFIC HEAT METHOD OF INVESTIGATING ATOM-WEIGHTS. 199 
 
 elements by reference to their Specific Heats ; i. e., to the number 
 of units of heat which proportionate weights thereof absorb in 
 acquiring a given sensible temperature (the unit of heat being 
 so much as suffices to raise 1 litre of water 1 degree centigrade). 
 Now it has been found that, though equal weights of the elements 
 absorb very different quantities of heat in attaining to equal 
 sensible temperatures, and therefore possess very unequal specific 
 heats, as thus measured ; yet if, of each element compared, a 
 quantity equal to its atom-weight be taken, the absorption of 
 heat, in attaining the same temperature, becomes precisely equal 
 for all ; or, to express the same thing in other words, the atom- 
 weights of the elements, multiplied into their specific heats, give 
 a constant product. This law is subject, unfortunately, to ex- 
 ceptions ; otherwise it would afford a very simple and universally 
 applicable mode of ascertaining the atom-weights of elementary 
 bodies. Even as it is, most valuable preliminary hints for 
 guidance, and welcome corroborations of chemically-obtained 
 results, are frequently afforded by these physical determinations ; 
 and the case of sodium, dwelt on above, may be cited as one of 
 those, in which the specific-heat investigation confirms the 
 indication of the atom-weight, derived from purely chemical 
 research (comp. p. 192). 
 
 It is remarkable that the specific-heat method often affords 
 the most distinct indications as to the atom-weights of ele- 
 ments, precisely in the cases in which such assistance is most 
 needed ; as, for instance, in the cases of the non-volatile elements, 
 referred to above ; and also in cases which, notwithstanding 
 that volatile compounds of the elements whose atom- weights 
 are sought, exist and have been weighed, are rendered doubtful 
 by the uncertainty whether one, two, or more of the atoms in 
 question are contained in the respective product-volumes. 
 Uncertainty of this kind prevailed, as we have pointed out 
 (comp. p. 194), in the cases of mercury, bismuth, and tin, until 
 it was set at rest by reference to their specific heats ; which 
 confirmed the view that the product-volumes of their chlorides 
 contain but one atom of the respective metals. There is evi- 
 
200 ISOMOEPHIC INVESTIGATION OF DOUBTFUL ATOM-WEIGHTS. 
 
 dently still greater doubt to set at rest in cases where the only 
 volatile bodies available for the determination of the required 
 atom- weights, by the vapour-density method, form series, whereof 
 all the members contain the same elements united in various 
 multiple proportions. In these cases also, two or more possible 
 atom-values offer themselves for selection, often with very 
 evenly-poised claims to preference; and it is then that the 
 determination of the specific heat comes in, with welcome 
 weight, to turn decisively the otherwise doubtful balance. 
 
 The phenomena of Crystallization afford similar assistance, 
 in a still larger number of doubtful cases, by revealing between 
 bodies certain chemical analogies which, in their turn, elucidate 
 questions as to their atom-weights. We shall hereafter learn 
 that bodies of similar atomic constitution are very commonly 
 isomorphic, i. e., that they affect similar, or closely-related, 
 crystalline forms. Hence the comparison of two crystalline 
 bodies, one of doubtful, the other of well-ascertained atomic 
 construction, will often afford conclusive information as to an 
 atom-value else uncertain. 
 
 It forms no part of our present plan to enter into the details 
 of these physical modes of research, nor to exemplify their 
 application to particular cases. Indeed, we have not yet 
 studied a sufficient number of the elements and their com- 
 pounds to prepare us for pursuing this line of research ; which 
 we shall have better opportunities of taking up hereafter, when 
 further advanced in our studies. 
 
 For the present we may close this branch of our inquiry by 
 subjoining a general table of the 61 elements, with their names 
 alphabetically arranged in the first column ; their symbols and 
 coefficients of quantivalence, with their atom-weights and volumes 
 in the second ; and their molecular symbols, weights, and volumes, 
 in the third. 
 
 The molecular values, and all the volumetric values, atomic 
 as well as molecular, are, it will be observed, given in this table 
 for the volatile elements only ; and not for all even of these. 
 The blanks await filling up with the results of future experi- 
 
EXPLANATION OF THE TABLE OF ELEMENTS SUBJOINED. 201 
 
 ment ; let us hope at no distant date. It is further to be ob- 
 served, with reference to this table, that a certain number even 
 of the atom-weights therein assigned, especially to the rarer 
 and less perfectly studied elements, are still matters in question 
 among chemists. These doubts are most prevalent in the cases 
 (already alluded to) of elements which combine in multiple pro- 
 portions to form series of compounds, from which this or that 
 may be chosen, at pleasure, to serve as the basis of the atom- 
 weight determination. In the present transitional state of che- 
 mistry, indeed, many of the atom-weights must be taken as 
 provisional, and subject to verification. In studying the indi- 
 vidual elements, we shall take occasion to set forth the experi- 
 mental facts, and the considerations founded thereon, which 
 have led to the adoption of each of the figures given in the 
 table ; and, perhaps, as we proceed, it may fall within our power 
 to correct some of the doubtful values, and to fill up some of the 
 blanks. These remarks apply not only to the molecule-forming 
 weights, but also to the coefficients of quantivalence by which 
 their atom-fixing powers are expressed. 
 
 The use of these last-mentioned exponents may be once more 
 mentioned in connection with this table. 
 
 In order to ascertain what weight of each element suffices to 
 fix or replace one standard univalent atom (H = 1), or a pro- 
 portionate part of any atom of higher quantivalence, we have 
 only to divide the atom-weight of any body by its quantiva- 
 lential coefficient. 
 
 The atom- weight of sulphur, for example, is shown, in column 
 3, to be 32 ; while its coefficient of quantivalence is given in 
 
 the column of literal symbols as S n . Therefore the quantity of 
 
 oo 
 sulphur requisite to replace 1 of hydrogen is =16 
 
 2i 
 
 With these explanations as to its employment, and subject to 
 these reserves as to its accuracy, this table will be found, I trust, 
 to convey, in a simple and compendious form, a considerable 
 mass of useful information. 
 
 K3 
 
202 
 
 TABULAR EPITOME OF RESULTS. 
 
 
 
 
 o a v w 
 
 
 
 
 
 ^- o 5 ^ -^ 
 
 s. 
 
 
 
 
 M rt t3 tiC 
 
 \. Cfi 
 
 
 
 
 a S ^ .2 -2 *S S 
 
 p g fl fe g o 
 
 .W. T< 
 
 
 
 
 ,J ^ 55 Q O 
 
 HrH \ 
 
 
 
 
 2 So S '5 rt "o <u 
 
 N. ^ 
 
 
 
 1 
 
 
 .. ._ ^\ 
 
 
 
 tk 
 
 m ^ 
 
 
 
 w* 
 
 | 
 
 B J '"S'S 5 
 
 <N 
 
 
 
 
 
 O 
 
 CO 
 
 
 O s 
 
 w 
 
 ^ 
 
 ^g-S g; 
 
 
 
 ^ 
 
 1 
 
 S ^ g 
 
 
 pmbolizatioi 
 
 
 naetric Symbolh 
 Elements 
 
 *1* 
 
 S J 8 
 
 If-IJi 
 
 n > @ u e 
 
 ^^0^ 
 
 w | 
 
 i 
 
 
 \ 
 
 111 
 
 
 
 
 ^ 
 
 2 fl bc^ ' 
 
 
 Q 
 
 
 g 
 
 rvf" "efi *^ C ^ S n 
 
 
 + 
 
 
 
 S S * & > 8 13 
 
 rv 
 
 'o 
 
 
 
 
 W 1 w\ 
 
 1 
 
 
 
 ^Il^l'^^ 
 
 
 iz; 
 
 1 
 
 ATOMS. 
 
 bolization 
 .ements. 
 
 ' ^ ^ - w> Ji 
 
 
 
 T^ t>* O^ l-O t>* ^d^ 
 
 cq CM i>. co TH 
 
 r-l i I 
 
 
 
 Sw 
 
 !i ^ 
 
 
 OQ 
 
 
 &*> 0) 
 
 1 
 
 
 i 
 
 
 IK 
 
 J ^ a .fe . 
 2 -3 -5 1 S 
 
 H H a 
 
 1 
 
 
 1* 
 
 g J a ffl 1 J 
 
 &-! ^!b o ^ o 
 <H GQ <H PQ PQ 
 
 w 
 w 
 
 
 
 
 
 
 
 
 1 
 
 .S 
 
 
 
 
 X 
 cj3 
 
 u - 
 
 
 
 
 
 B 'S 
 
 K 
 
 
 t/2 
 
 1 
 
 1 g 
 
 C3 cS 
 
 
 
 A 
 
 g 
 
 
 "^ ' ' 
 
 
 
 O> 
 
 4> "S 
 
 
 
 * 
 
 1 
 
 ! i 
 
 ll I g i 
 
 
 
 O 
 
 
 5S *d g 'g 1 
 
 
 
 
 
 S P5 *= QD S &* 
 Jj r 1 ^1 <S O 
 
 
 
 
 
 
TABULAE EPITOME OF RESULTS. 
 
 203 
 
 o 
 3 
 
 Q < 
 
 oo o 
 
 O 00 
 
 10 
 
 10 oo 10 
 
 <M OO CO 
 O O O 
 
 CO Ci O 
 O5 TH C5 
 
 co c3 g 
 
 O Q O 
 
 3 S I 
 
 M M 
 
204 
 
 TABULAR EPITOME OF EESULTS. 
 
 
 
 c 2 . 
 
 
 
 
 -'p .2 ^ "5 -S 
 
 
 
 s 
 
 Wt! 1 !! 
 
 a - -a 
 
 
 Q 
 
 o "S 
 
 
 
 2 
 
 ^ ^ 
 
 
 3 
 
 o 
 
 QJ g _C 
 
 b 2 .* | -S. 
 
 
 a 
 
 ,Q 
 
 
 O^ ^^ 
 
 j 
 
 "*-* 
 
 O o ^ Jg ^ 
 
 iO 
 
 3 
 
 
 W 
 
 C ' 
 
 S?li|j 
 
 
 o 
 
 'is 
 
 s *" S 
 
 
 S 
 
 $ 
 
 (4-1 
 
 
 
 J H 
 
 <o ^* 
 
 
 
 s a 
 
 2 S^ 
 
 
 
 rs 
 
 .2 
 
 ilfi! 
 
 CO **-s o ts 
 
 W HH 
 
 w 
 
 
 s 
 1 
 
 g-'ll 
 
 
 
 ~g 
 
 18 
 
 
 
 "- ' 
 
 a 
 
 
 
 o 
 
 "S ~ =T J M i ~ 
 
 
 
 CH 
 
 | | | || || 
 
 W M 
 
 
 
 ^f Hl^ 
 
 
 05 
 
 S , 
 
 
 
 ^ bfl ^i 
 
 
 g ( 
 
 S a 
 
 3 .sf "5 i-2 'c '^ 
 
 iH 00 t- CM t- 
 
 Ci CM O O 
 
 <J 
 
 2 1 
 
 J "1 o. 1 S 
 
 tH rH CM 
 
 
 HS 
 
 rt 8 S 
 
 
 
 
 43 > 
 
 
 
 "F -^ 
 
 O "3 -.j .2 '-^ a ; 
 
 > M G K 
 
 
 01 S 
 
 g | g | || 
 
 
 
 I ^ 
 
 e 
 
 ' 
 
 
 
 a 
 
 - 
 
 
 cS 
 
 1 J" 
 
 
 CO 
 
 -2 
 
 i s 
 
 
 cy 
 S 
 
 g 
 
 CS g 
 
 . 
 
 << 
 
 
 
 ~ "o 
 
 ci 
 
 SB 
 
 Ed 
 
 "-2 ^ 
 
 3 
 
 
 a> 
 
 J* 
 
 3 g 
 
 
 
 1 = 
 
 Ii 1 ill 
 
 K M HH M h^ M 
 
 1 
 
TABULAR EPITOME OF RESULTS. 
 
 205 
 
 to 
 
 w 
 
 to 
 
 525 ' 
 
 00 
 
 CO 
 
 to 
 
 W 
 
 8 
 
 O d CO b- 
 
 O C5 O C5 
 
 d 
 
 i I 05 T-l O CO 
 
 M 
 
 
 
 
 3 a 1 K 1 g 
 
 ^5 fe O O PM 
 
 i 
 
 & 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ' 1 i ' 1 : 
 
 3 Ss fl 
 
 1 8 I I 1 1 ^ 
 
 ill 1 ll 
 a a a a g 
 
 g g a - '* I 
 
 1 I ill 
 
 cc 
 
 i 
 
 l 
 
 Platinum . . 
 Potassium . . 
 
206 
 
 TABULAR EPITOME OF RESULTS. 
 
 
 
 o Cr g 
 
 
 
 
 jf^fc,f i 
 
 
 
 
 s g 1 s g -i 
 
 GO 
 
 
 
 
 111 11 !^ 
 
 C^ 
 
 
 
 
 s gcr ! 
 
 
 
 r-o 
 
 M 
 
 
 
 !*> 
 
 0, t JS 
 
 
 CO 
 
 I 
 
 H ^ kio 
 
 CO 
 
 10 
 
 S 
 
 
 
 g;l ^1 
 
 rH 
 
 o 
 
 in 
 
 s ^'s 
 
 
 
 ii i 
 
 B *s ? 
 
 
 
 P 
 
 J^ll 
 
 
 
 o 
 
 Oil! 
 
 
 
 O> 
 
 Pjs I " 8 
 
 
 
 
 co ~ ,2 
 
 
 
 1 
 
 111 
 
 
 
 i 
 
 3 
 
 
 
 1 
 
 H -ll^si 
 
 
 
 
 ||||||| 
 
 <2 
 
 
 
 " - -5 
 
 
 ATOMS. 
 
 ll 
 ^ s 
 
 gS^|l^. 
 
 ^H XO -rH O5 CO CO CO 
 
 o oo o t- o cq cq 
 
 
 ji.2 
 
 ^ 4J tt P-l QJ ^t-l ^ 
 
 
 
 ?: y 
 
 ^ ^ -^ 
 
 
 
 j8 ' 
 
 ^> eS 
 
 
 
 
 iJ C^ c .^ . 
 
 
 
 SI 
 
 '11 g 1 
 
 rd !2a M tf ^ M ao . "es 
 
 
 la ^o 
 
 ^ "S * W 4-^ 
 
 PH P^ Pn GO <1 GO *3 
 
 
 fc 
 
 o ^5 
 
 
 
 
 
 
 
 
 
 
 
 TJ 
 
 S 
 
 
 
 X 
 
 'g ^ 
 
 
 
 U3 
 
 DA 0) 
 C T3 
 
 
 a 
 
 1 
 
 t< O 
 
 
 < 
 
 1 
 
 1 1 
 
 g 
 
 
 1 
 
 (4-1 
 O 
 
 "o -^ 
 
 13 
 a 
 a 
 3 
 
 1 1 1 i 1 1 1 
 
 Ps PH CO CO bo CO 
 
 1 
 
 1 
 
 I 
 
TABULAR EPITOME OF RESULTS. 
 
 207 
 
 = S B S d ^ 
 P f> i* N tsi 
 
 Strontium 
 
 I l| I I 
 
 I 1 4 1 I .a .1 
 
 & H H H H H H 
 
208 NITROGEN-SYMBOL APPEAL THEEEON TO FKANCE. 
 
 One conspicuous merit of the elementary symbolization epi- 
 tomized in this table consists in its Universality ; I mean in its 
 general recognition and employment by the chemists of all 
 nations throughout the civilized world. 
 
 From this broad statement, however, we have to except one 
 symbol, and one nation. That symbol is the expression N = ni- 
 trogen ; and that nation is France ; for, France alone, among all 
 the nations of the earth, continues to designate nitrogen by tho 
 symbol and name Az = azote. It is time that this single devi- 
 ation from uniformity in the elementary chemical language of 
 the world should disappear ; and I confidently appeal to our 
 scientific brethren beyond the Channel to efface this one remain- 
 ing discrepancy by adopting the symbol N. 
 
 It is not necessary, for this purpose, to discontinue the use of 
 the name azote also. The Italians employ the term azoto, but they 
 are beginning to couple with it the symbol N. In the cases of the 
 alkali metals, the French themselves concur with other nations 
 in using the symbols K and Na, though these are not the initial 
 letters of the names (jpotasse and soude), which they assign to the 
 alkalies. Evidently, therefore, there would be no logical incon- 
 sistency involved in this slight concession, on the part of France, 
 to the general usage of the scientific comity of nations. On the 
 contrary, France owes this rectification quite as much to her 
 own genius for philosophical order and unity, as to that general 
 European sentiment whereof, in this appeal, I am but the humble 
 spokesman. Moreover, to a similar, and much more sweeping 
 appeal, on the part of France, scientific Europe has responded with 
 alacrity, by renouncing, in favour of the noble French metrical 
 system, a thousand incongruities of local weight and measure. 
 And again, in the great work of building up, on a unitary basis, the 
 magnificent edifice of modern chemistry, France has taken, in 
 the past as in the present, her fully proportionate share. These 
 are all valid and logical arguments in support of our present 
 plea. When, therefore, we solicit at her hands the abandon- 
 ment of her one discordant symbol, we do but ask her to carry 
 out the spirit of her own unitary conceptions ; and to the suc- 
 cessors of Lavoisier to the contemporaries of Gerhardt our 
 appeal, we are persuaded, will not be made in vain. 
 
209 
 
 LECTUKE XII. 
 
 Compounds of a higher order, ternary, quaternary, &c. laws of proportion- 
 ality and quantivalential relations exemplified in the generation of such 
 compounds their frequently high vapour-densities their tendency, in 
 many cases, to dissociation examples of ternary compounds hydrochlo- 
 rate of ammonia its production by the union of the molecules of two 
 gaseous binary compounds its neutral, salt-like characters dissociation 
 of its vapour ternary compounds produced during the progressive 
 dehydrogenation of water and ammonia by the alkali-metals uni- 
 sodic water, or hydrosodic oxide unisodic and bisodic ammonias, or 
 hydro-unisodic and hydro-bisodic nitrides progressive expulsion and 
 replacement of hydrogen- atoms by sodium-atoms during the formation of 
 these compounds analogous substitution-compound in the marsh-gas 
 group general conception of substitution-compounds their production 
 further illustrated in the progressive transformation of water, ammonia, 
 and marsh-gas, by inception of chlorine and expulsion of hydrogen reten- 
 tion of the structural type of the parent compound by its substitutional 
 derivatives conversion of binary into ternary compounds by atomic incep- 
 tion, unattended by substitutional displacement exemplification of this 
 mode of their genesis in the hydrochloroxygen series analogous ternary 
 compounds of the hydrochloric group of the water group hydrosulph- 
 oxygeu series of the ammonia group hydrophosphoxygen series of the 
 marsh-gas group methylic alcohol its transitional importance close 
 of the lecture summary recapitulation and conclusion of the course. 
 
 THE principles of symbolization, literal and diagrammatic, illus- 
 trated in the table of the elements to which our attention was 
 directed at the close of the last lecture, apply equally to the 
 concise representation of the structure of compound bodies. 
 
 Thus, as, in the table, we denote the elementary hydrogen 
 molecule by the letters HH, and depict diagrammatically its 
 diatomic structure by the dilitral figure 
 
210 
 
 COMPOUNDS OP A HIGHER ORDER ; THEIR OBEDIENCE 
 
 so, likewise, we represent the similarly-formed diatomic com- 
 pound molecule of hydrochloric acid by the letters HOP, and 
 the dilitral figure 
 
 Again, as in point of structure and dimensions the polyato- 
 mic binary molecules of water-gas, ammonia-gas, and marsh-gas, 
 are all precisely analogous to the elementary diatomic standard 
 molecule, HH, we represent these also, respectively, in the lite- 
 ral expressions H 2 n , H 3 N m , and H^, and by the dilitral double- 
 square symbols 
 
 When we wish to represent elementary or compound molecules, 
 whose dimensions have not been ascertained by experiment, but 
 are provisionally assumed on analogical grounds, we resort to the 
 dotted lines previously employed by us in like cases ; represent- 
 ing, for example, the hypothetical gaseous molecules of the 
 alkali-metals, and their compounds with chlorine and oxygen, as 
 follows : 
 
 Na'Na 1 
 
 K'K 1 ! and i Na'Cl 1 
 
 K'CP 
 
 .J L. 
 
 Though we have hitherto had occasion to employ these 
 methods of symbolization only for the representation of the ele- 
 ments and their binary compounds, they are equally available 
 (mutatis mutandis) to denote the composition and structure of 
 compounds of a higher order ; that is to say, of compounds in 
 whose formation not two only, but three, four, five, and some- 
 times even six elements take part, and which are accordingly 
 distinguished as ternary, quaternary, quinternary, and sexternary 
 compounds. 
 
TO THE LAWS WHICH GOVERN SIMPLER COMPOUNDS. 211 
 
 Of these compounds of a higher order we have as yet re- 
 frained from speaking ; though, in the course of our experiments, 
 even of the very first we made several such bodies passed 
 through our hands. The fact is that, in those earlier stages of 
 our inquiry, reference to these bodies would have been prema- 
 ture and purposeless ; a mere incumbrance of the memory with 
 details not contributing to the elucidation of the subjects then in 
 hand. The time, however, is now come when we may, with 
 advantage, pick up these dropped links and add them to our 
 chain. The study of their more complex structure will prepare 
 us for the investigation, upon which we must hereafter enter, of 
 an almost infinite succession of complex bodies, so numerous 
 and so ever multiplying, that he who desires to be a master in 
 chemistry must be content to remain, his whole life through, a 
 hard-working student of their legions. 
 
 We may premise briefly, with respect to these compounds of 
 a higher order, that they are generated by the operation of the 
 same forces, exercised in accordance with the same laws, as deter- 
 mine the production of the binary compounds. The activity of 
 those forces is as often attended, in the one as in the other class 
 of cases, by perturbations, manifested in the development of light 
 and heat. The laws of proportionality are, moreover, as exactly 
 fulfilled in the formation of the most complex of these higher 
 compounds as in that of the simplest molecule ; each atom that 
 takes part in a combination invariably engaging therein either 
 in the single or multiple ratio of its weight, and carrying with 
 it unchanged, throughout the most complicated reactions, its 
 peculiar atom-fixing or quantivalential powers. 
 
 Hence it results that, in the formation of many of these com- 
 pounds of a higher order, the dilitral product-volume, into which 
 we have as yet seen condensed only some half dozen volumes at 
 most, becomes packed, so to speak, with scores of volumes. 
 Hence we may readily conceive that, in the study of the more 
 complex, as of the simpler bodies, the "product-volume is still 
 (whenever they are volatile) the starting-point of the investi- 
 gation ; that is to say, we begin by determining the ponderal 
 
212 NATURE AND EFFECTS OF DISSOCIATION 
 
 composition and vapour-density of the body under examination, 
 and thence deduce the weight and constitution of its dilitral 
 volume, which corresponds, as we know, with the weight and 
 structure of its molecule. 
 
 This method is, however, subject in all cases to a difficulty 
 on which we have not yet touched, and which it is now oppor- 
 tune to point out and illustrate. 
 
 It frequently happens that our attempts to determine the 
 vapour-densities of volatile compounds, whether binary or of a 
 higher order, are baffled by their splitting up, at the temperature 
 of vaporization, into their constitutent gases or vapours ; which, 
 of course, how much soever they may have been condensed 
 during combination, resume, in the act of separating, their 
 proper volumes. 
 
 We need not look far, either for examples of compounds of a 
 higher order, or for illustrations of this splitting up, or, as it is 
 termed, Dissociation, of volatile bodies, under the influence of 
 elevated temperatures. 
 
 Among the experiments we have made here together, you 
 doubtless remember one in which, when acting on ammonia with 
 chlorine gas (comp. p. 56), we incidentally produced a white 
 crystalline deposit on the interior of the tube we employed. 
 That deposit we laid aside at the time, as having no bearing on 
 the question then under consideration, viz., the volume-ratio of 
 the two gaseous constituents of ammonia. But now this deposit 
 acquires particular interest for us, firstly, as being a ternary 
 compound, and, secondly, as exemplifying the phenomenon of 
 dissociation. 
 
 We will, therefore, fix our attention on this body ; and, for 
 this purpose, we will begin by producing a small quantity of it ; 
 a result which we may readily accomplish by simply bringing 
 its constituent gases into contact with each other. 
 
 With the experience in manipulation we now possess, we can 
 be at no loss to perform this easy operation. Here is our 
 familiar pneumatic apparatus, our mercurial trough, with its in- 
 verted cylinders, for the reception and admixture of the gases 
 
AS EXEMPLIFIED IN THE CASE OF AMMONIA HYDROCHLORATE. 213 
 
 under examination. A pair of these cylinders, of equal size, 
 have been already filled, by the means we have so often employed, 
 the one with dry ammonia-gas, the other with dry hydrochloric- 
 acid gas. Beside these receptacles, a third, of larger size, 
 stands inverted over the trough ; but this one is filleci to the top 
 with mercury, in readiness for the reception of gas. Into this 
 cylinder we first pass, by upward decantation, the hydrochloric 
 acid gas (marking the space it fills by a caoutchouc ring), and 
 then, bubble by bubble, we add the ammonia. 
 
 FIG. 65. 
 
 As each bubble of this gas comes into contact with the hydro- 
 chloric acid, a dense white cloud is seen to form ; and this con- 
 denses into crystalline flakes, which collect on the interior of the 
 vessel. During this process the original volume of the hydro- 
 chloric acid gas gradually diminishes. Indeed, the more ammonia 
 we pour up through the mercury, the higher does this metal rise 
 in the cylinder ; and with the last bubble of ammonia sent up, 
 the last trace of the gaseous mixture disappears, and the vessel 
 now only contains, beside mercury, the film of white deposit 
 into which the two gases have condensed. 
 
214 CONSTITUTION OF AMMONIA HYDROCHLORATE. 
 
 We are, therefore, justified in regarding this deposit as a com- 
 pound of the two gases we have mixed, viz., hydrochloric acid 
 and ammonia, united in equal volumes. 
 
 As a corroborative demonstration of this fact, we have here 
 the means of mixing the two gases as before, but with excess of 
 either gas at pleasure. Into this cylinder we introduce hydro- 
 chloric acid gas in excess of the ammonia ; into this other one, 
 ammonia in excess of the hydrochloric acid : in both cases, you 
 observe, combination takes place as before ; but in both cases 
 there is a surplus of uncombined gas left. By means of these 
 test papers we further learn that the surplus consists, in one 
 case, of ammonia ; in the other, of hydrochloric acid a clear 
 proof that combination has only taken place between equal 
 volumes of the two gases. 
 
 The white, solid product we find, upon trial, to be readily 
 soluble in water ; and, on comparing its solution with solutions 
 of its respective constituents, we find that the characteristic 
 properties of these latter have entirely disappeared in the pro- 
 duct of their union. The suffocating exhalation of the hydro- 
 chloric acid, the pungent odour of the ammonia, are no longer 
 perceptible in the solution of the compound they have formed. 
 This latter solution does not, like the hydrochloric acid solu- 
 tion, redden litmus-tinted paper ; nor does it, like the ammonia 
 solution, restore the blue colour to litmus-paper reddened by 
 acidulation. With two bodies of strongly marked, and decidedly 
 opposite characters, we have formed a neutral compound, in 
 aspect and general characters strongly resembling common culi- 
 nary salt. This product, in conformity with a very usual artifice 
 of chemical nomenclature, is distinguished by a name derived 
 from the appellations of its two constituents ; it is called, at 
 pleasure, either ammonia-liydrochlorate, or hydrochlorate of am- 
 monia. 
 
 This salt-like body is readily volatilizable, and were it not 
 subject to the phenomenon of dissociation, to which our atten- 
 tion has already been directed, we should have no difficulty in 
 establishing its normal molecular coostitution as follows : 
 
OTHER TERNARY COMPOUNDS. 215 
 
 Hydrochloric Add. Ammonia. Ammonia-liydroclilorate. 
 
 1 molecule. 1 molecule. 1 molecule. 
 
 The dimensions of the molecule are here marked in dotted 
 outline, our sign of doubt. But of the ponderal constitution of 
 this compound we are in no doubt at all. We know that this 
 product is a ternary compound, consisting of hydrogen, 4 vols., 
 united with nitrogen and chlorine, one volume each. If, as we 
 assume, these 6 volumes of constituent gases condense into the 
 normal dilitral product-volume, we have here a condensation 
 ratio of f = 1. 
 
 But, while we provisionally adopt this view, we must carefully 
 bear in mind that it rests upon analogy only, not on fact. Every 
 attempt hitherto made to obtain the assumed normal vapour, by 
 volatilizing hydrochlorate of ammonia, has resulted in the pro- 
 duction of a quantity of gas equal to twice the normal product- 
 volume. In other words, the volume obtained is equal to the 
 sum of the volumes of the two constituent gases, in their free 
 state. These are accordingly believed to dissociate during the 
 act of vaporization of the ammonia-hydrochlorate. 
 
 But our early experiments supply us with other examples of 
 ternary compounds, which call in their turn for brief considera- 
 tion here. 
 
 In studying the composition of hydrochloric acid, water and 
 ammonia, we employed, as you remember, an alkali-metal, pre- 
 ferably sodium, to liberate the hydrogen from those several 
 compounds ; the final result being, in each case, that the 
 sodium replaced the hydrogen set free, and combined in its 
 stead with the chlorine of the hydrochloric acid, with the 
 oxygen of the water, and with the nitrogen of the ammonia. 
 
 Now, as chlorine, oxygen, and nitrogen are respectively uni-, 
 bi-, and tri-valent ; the molecules of their hydrogen compounds 
 contain respectively (as we well remember) one, two, and 
 three atoms of hydrogen. 
 
216 TERNARY PRODUCTS OF THE DECOMPOSITION OF WATER 
 
 Evidently, therefore, from the 'molecule of hydrochloric acid, 
 which only contains one atom of hydrogen, either all or none of 
 the hydrogen must be displaced. 
 
 From the molecule of water, on the contrary, containing as it 
 does two atoms of hydrogen, it is possible to conceive the dis- 
 placement of those atoms by two successive invasions (so to 
 speak) of the alkali metal ; one atom, or half the whole quantity, 
 being first displaced, and afterwards the remainder. 
 
 Again, from the molecule of ammonia, with its three atoms 
 of hydrogen, we readily understand that sodium may succes- 
 sively displace first one, and then a second, of those hydrogen 
 atoms, before finally displacing the third. 
 
 Now, though, at the time of making these experiments, we 
 were only interested in, and therefore only dwelt on, the final 
 result, viz., the complete expulsion of the hydrogen from these 
 three compounds, and the consequent formation of binary com- 
 pounds only between the sodium and the chlorine, the oxygen, 
 and the nitrogen respectively ; it now becomes opportune to 
 dwell on the fact that, in the cases of water and ammonia, the 
 final result is not attained at once, but that the above-mentioned 
 intermediate ternary compounds are in reality produced during 
 the successive stages of the process. 
 
 The ternary products of the progressive dehydrogenation of 
 water and ammonia by sodium are, in the diagram subjoined, 
 placed intermediately, between the original compounds, water 
 and ammonia, which are figured on the left, and the final pro- 
 ducts, oxide and nitride of sodium, which appear to the right. 
 For clearness' sake the atomic constitution of each product is 
 displayed without condensation. 
 
AND AMMONIA BY SODIUM. 217 
 
 Progressive Decomposition of Water and Ammonia. 
 
 ORIGINAL. 
 (Binary}. 
 
 INTERMEDIATE. FINAL. 
 (Ternary}. (Binary). 
 
 Na 1 ! 
 L j 
 
 JNa'j 
 
 
 H 
 
 
 0" 
 
 i ; 
 
 O 11 
 
 O n 
 
 
 H 
 
 H 
 
 
 jNa 1 ! 
 
 
 (hypothetical) 
 
 H 
 
 N 
 
 Na 1 ; 
 
 Na 1 
 
 The one ternary compound here shown between water and 
 soda consists, as you observe, of one atom each of sodium, 
 hydrogen, and oxygen. It may be considered as water, in 
 which half the hydrogen is replaced by sodium. It is a sub- 
 stance well known to us all, being the common hydrated or 
 caustic soda, so commonly employed in the arts and manufac- 
 tures. The final binary product, from which all the hydrogen 
 has disappeared, sodium alone remaining in combination with 
 oxygen, only retains, of the parent-compound water, from which 
 it sprung, the characteristic structural type. It is, therefore, 
 often called sodic anhydride, meaning literally water-less soda ; 
 for which the term dry oxide of sodium is often substituted. 
 These remarks, as well as the diagrammatic representations them- 
 selves, are as applicable to potassium as to sodium, and may be 
 converted into potassic expressions by merely substituting K 
 for Na throughout. Or the initial M (for metal) may be used 
 to give the expressions a general form, coinclusive of both the 
 
 L 
 
218 PROGBESSIVE INCEPTION OF SODIUM BY WATER AND AMMONIA. 
 
 alkali-metals an artifice of symbolic generalization very com- 
 monly employed by chemists. 
 
 In the. ammonia-sodic series, figured below the aqua-sodic 
 series in the diagram, we see two ternaries, one unisodic, the 
 other bisodic, intervening between the unmodified ammonia and 
 the nitride of sodium which is the final product of its transfor- 
 mation. The bisodic compound is hypothetical. 
 
 The reactions by which, in the case of water and ammonia, 
 these serial products are obtained, each in succession by the 
 enrichment of a previous one with additional sodium, are dis- 
 played in the following molecular equations, which also show 
 the proportion of hydrogen liberated at each stage of the opera- 
 tion : 
 
 1. Progressive Inception of Sodium by Water. 
 
 1st Stage ..2 
 
 H 2 0" 
 
 + i Na'Na' j = 2 
 
 N^HO" 
 
 
 
 
 Na'HO 11 
 
 -f-j Na'Na 1 j = 2 
 
 Na^0 
 
 2. Progressive Inception of Sodium by Ammonia. 
 
 1st Stage ..2 
 
 3rd Stage ..2: 
 
 H 3 N m 4- 
 
 Na'Na* 
 
 = 21 Na^N 
 L _j_ 
 
 
 
 
 Na'H 2 N' -j 
 
 Na'Na 1 
 
 = 2! NajHN" 1 
 
 
 
 
 Na^HN m '|-f- 
 ._____ .J 
 
 Na'Na 1 
 
 = 2! Na 3 N m 
 
 These equations bring pointedly under our notice the re- 
 markable fact, that, in the progressive transformation of ammonia 
 by sodium, each additional atom of metal that enters into the 
 compound takes the place of a hydrogen-atom ; which is, as it 
 were, expelled to make room for it. 
 
GENERAL CONCEPTION OF SUBSTITUTION-COMPOUNDS. 219 
 
 Our fourth typical group, that headed by marsh-gas, affords 
 us one case, and one only, of a ternary compound exempli- 
 fying this peculiar mode of formation ; that is to say, resulting 
 from the displacement of a hydrogen-atom by a sodium-atom 
 entering the molecule in its stead. The formula of this ternary 
 compound is 
 
 This product is formed with considerable difficulty, and by 
 circuitous processes, upon which we cannot here dwell. We 
 refer to it only for the sake of remarking tha't each of our four 
 typical groups exemplifies the formation of ternary com- 
 pounds by the expulsion of hydrogen, and the replacement of 
 the atom or atoms so expelled by a quantivalent proportion 
 of sodium. 
 
 By these particular examples we are led up to the general and 
 most important conception of Substitution-compounds ; that is to 
 say, of bodies formed (often in extensive series) by the replace- 
 ment of one or more of the constituent atoms of a compound by 
 atoms of some other body introduced in their stead. And we 
 thus make acquaintance, in germ, with a principle, from which, 
 as from a living seed, the mighty growth of modern chemistry 
 has mainly sprung. 
 
 This principle is too important to be passed over lightly : it 
 deserves the best elucidation that the experiments which we have 
 already tried together enable us to bestow on it. 
 
 Among our early experiments on water, ammonia, and marsh- 
 gas, we remember effecting their decomposition by employing 
 chlorine to withdraw from them their hydrogen atoms, so as to 
 liberate from water oxygen, from ammonia nitrogen, and from 
 marsh-gas carbon. (Comp. pp. 28, 33, and 102.) 
 
 At that stage of our inquiry it would have served no useful 
 purpose to point out that, under certain conditions of procedure, 
 the withdrawal of hydrogen from those compounds by chlorine 
 takes place in the same progressive manner as we have seen it 
 
220 FORMATION OF SUBSTITUTION COMPOUNDS, 
 
 affect when sodium was the displacing agent employed. This 
 fact now interests us deeply as a corroboration of the great law 
 of substitution, which is dawning on our view. 
 
 The reaction in question takes place in the following manner. 
 While part of the chlorine is withdrawing hydrogen from water, 
 ammonia, or marsh-gas, as the case may be, and forming hydro- 
 chloric acid with that displaced hydrogen, another portion of 
 chlorine takes the place of the said hydrogen in the compound 
 from which it is expelled, the final products of the transforma- 
 tion being 
 
 In the case of water* 
 
 C1^0 n , instead of H 2 n . 
 In the case of ammonia 
 
 Cl^N 111 , instead of H 3 N" r . 
 In the case of marsh-gas 
 
 C1^C IV , instead of H 4 C n '. 
 
 Evidently, here also, as in the decomposition by sodium, we 
 may expect the formation of intermediate ternary compounds, 
 due to the partial and progressive replacement of the hydrogen 
 atoms by atoms of the attacking agent. 
 
 In the following diagrams we have (as before) placed between 
 the original and final binary compounds the ternary compounds 
 capable of intermediate production. 
 
 * Practically, the metallic derivatives of water are employed for this 
 purpose ; their decomposition by chlorine being more readily effected than 
 that of water itself. 
 
TERNARY CHLORINE SUBSTITUTION COMPOUNDS. 
 
 221 
 
 Compounds. 
 
 ORIGINAL. 
 (Binary}. 
 
 H 
 
 H 
 
 H 
 
 O n 
 
 C iv i 
 
 _ 
 
 { nn 
 
 INTERMEDIATE. 
 
 (Ternary]. 
 
 H 
 
 
 
 H 
 
 
 N m 
 
 
 Cl 1 
 
 
 N ra 
 
 FINAL. 
 (Binary.} 
 
 cr 
 
 
 N" 1 
 
 From this diagram we gather, at a glance, that between water 
 and the bichloride of oxygen there intervenes but one ternary 
 compound ; while, between ammonia and terchloride of nitrogen, 
 two ternaries find place, and between marsh-gas and tetrachloride 
 of carbon, no less than three ternaries are interposed. The 
 intermediate ammoniacal compounds have never yet, it must be 
 admitted (partly on account of their dangerously explosive 
 character), been produced in sufficient purity and abundance for 
 analysis; though, of their existence, analogy scarcely permits 
 
 L 2 
 
222 CHLORINE DERIVATIVES OF WATER, AMMONIA, AND MARSH-GAS. 
 
 a doubt. All the other compounds symbolized are bodies of 
 well-ascertained composition and properties. 
 
 The production of these substitution compounds, intermediate 
 and final, each from its predecessor, by additional inception of 
 the invading element, at the expense of hydrogen expelled and 
 converted into hydrochloric acid, may be shown as before by 
 ordinary equations; which fall, in this case, into three sets, 
 respectively representing the progressive transformation of 
 water, ammonia, and marsh-gas. 
 
 TERNARY CHLORINE-DERIVATIVES : 
 1. Of Water. 
 
 HC10 j 
 
 HN 
 
 H 2 C1N 
 
 HC1 2 N 
 
 + HC10 ! 
 
 4- 
 
 HC1N 
 
 HC1 2 N 
 
 C1 3 N 
 
THREE MODES OF TRANSFORMATION OF BINARIES INTO TERNARIES. 223 
 
 We have not yet, however, learnt from these diagrams all 
 that they can teach us. They shadow forth clearly another 
 and a most general fact of modern chemistry, viz., the uniform 
 retention, by substitutional derivative compounds, of the 
 structural type affected by their primary or parent compound. 
 
 We must not dwell here on this pregnant theme the law 
 of atomic substitution; nor, indeed, have we acquired facts 
 enough, as yet, to illustrate its nature, and prove its importance. 
 A moment's reflection will, however, enable us to conceive how 
 vast must be the aid derivable from such a principle in the ful- 
 filment of the task, else almost impracticable, of distributing 
 into a system of natural groups the ever-growing multitudes of 
 serial compounds. 
 
 From the foregoing remarks and examples, we have learned 
 that a binary compound molecule may become ternary, either 
 by uniting with another molecule (as in the case of the ammonia- 
 hydrochlorate), or by receiving into its structure one or more 
 atoms, of a third body in replacement of a proportion of its own 
 atoms. In this latter class of cases it is found that the number 
 of quantivalents brought in by the atoms introduced is always 
 exactly equal to the number of quantivalents carried away in the 
 atoms expelled. It is true that, in the examples given, we have 
 only employed univalent chlorine and sodium to replace, atom 
 for atom, univalent hydrogen. We shall, however, hereafter 
 learn that this law of substitutional quantivalence is universal, 
 and holds good in all cases of substitutional reaction, whether 
 the incoming and outgoing atoms be uni-, bi-, tri-, or quadrivalent, 
 or be partly of one class and partly of another. The law of 
 which we thus obtain a first glimpse is one of the most im- 
 portant results of modern chemical research, and it will be 
 continually pressed on our notice during the course of our 
 future studies. 
 
 There is yet a third manner, besides the two just mentioned, 
 in which a binary compound molecule may become ternary ; 
 and this consists in its direct assumption of one or more atoms 
 of a third element, without loss, by substitutional displacement, 
 of any atom or atoms of its own. 
 
224 HYDEOCHLOEOXYGEN AND HYDEOSULPHOXYGEN SEEIES. 
 
 Oxygen, for example, is thus taken up by the majority of our 
 typical binary hydrogen-compounds and their congeners. Hydro- 
 chloric acid, the type of our first group, may be cited as pos- 
 sessing this faculty in a high degree. Its molecule can take up 
 1, 2, 3, or 4 atoms of oxygen, and form therewith a series of 
 four well-marked ternary compounds, which may be called the 
 hydrochloroxygen series, and which are represented by the fol- 
 lowing expressions : 
 
 Hydrochloroxygen Series. 
 
 Hero*, HCPOS, HCPOS, Hero 1 ]. 
 
 The congeners of hydrochloric acid, hydrobromic and hydri- 
 odic acids possess similar properties, and both of them form 
 series of oxygen compounds, analogous to the hydrochloroxygen 
 series above. 
 
 This faculty of taking up oxygen is by no means wanting in 
 our second group of typical compounds, though it is more 
 conspicuous in the congeners than in the architype, water 
 itself. 
 
 Water, indeed, does form one compound, by the inception of 
 additional oxygen ; but the product is not easy to obtain, nor is 
 it of stable character. But the analogues of water the hydrogen 
 compounds of sulphur and selenium form, each of them, two 
 well-known ternary compounds, by incorporating, in their 
 respective molecules 3 atoms and 4 atoms of oxygen ; whence 
 we may reasonably presume that analogous intermediate combi- 
 nations of these bodies with 1 and 2 atoms of oxygen are 
 possible, and may one day be obtained. The entire series, 
 actual and possible, of the sulphuretted-hydrogen oxides is dis- 
 played in the subjoined set of expressions : 
 
 Hydrosulplioxygen series. 
 HS n O n 
 
 The two first formulae of this series represent its as yet 
 hypothetical members. The last two, on the contrary, represent, 
 with perfect accuracy, the composition of two of the most 
 
HYDROPHOSPHOXYGEN SERIES. METHYLIC ALCOHOL. 225 
 
 important acids known to chemistry, as well scientific as in- 
 dustrial, viz., the monohydrated sulphurous and sulphuric 
 acids. These also we must pass in our rapid course ; reserving 
 for a more fitting occasion the cultivation of their valuable ac- 
 quaintance. 
 
 Passing to our third typical group, headed, as we remember, 
 by ammonia, and comprising among its members phosporetted 
 and arsenetted hydrogen, we find that, though ammonia itself 
 has not yet been induced to combine with oxygen, the analogue 
 of ammonia, phosphoretted hydrogen, is conspicuous for the 
 number and variety of its ternary compounds with that body ; 
 combinations of the molecule H 3 P m with 2, 3, and 4 atoms of 
 oxygen having actually been obtained, so that only its possible 
 protoxide remains to be produced to complete the quadruple 
 series. The formation and composition of these ternaries, in- 
 cluding the as yet hypothetical member of the series, are dis- 
 played in the following expressions : 
 
 HydropJiosplioxygen series. 
 H 3 P ra O n , H 3 P m O n 2 , 
 
 Our fourth and last typical group, at the head of which stands 
 marsh-gas, has hitherto furnished only one example of a ternary 
 oxygen compound ; which has the formula H 4 C lv O n . This 
 compound, however, which we shall hereafter study under the 
 name of Methylic Alcohol, makes up. by its extreme importance, 
 for the else sterile character of the marsh-gas group, in respect 
 of ternary oxygen-derivatives. Through methylic alcohol, as 
 through a widely opened gate, we shall pass to the study of 
 its almost innumerable derivatives, which form in themselves 
 one of the largest tracts of that domain through which our 
 future journey lies. 
 
 But now, fellow-travellers, we have arrived at a point where 
 our progress together must be interrupted for a time. The 
 plan of this brief introductory course is fulfilled ; and its objects 
 
226 RECAPITULATION AND CONCLUSION. 
 
 so far as my limited powers permit their achievement, are 
 attained. Before we part, however, let us, as wayfarers are 
 wont, rest a little while on the gentle eminence we have 
 attained, and take a retrospective survey of our route thus far. 
 In other words and to drop metaphor let us endeavour, by a 
 summary recapitulation, to fix in our memory some of the 
 leading facts and principles which our experiments and reason- 
 ings, our inductions and deductions, have gradually unfolded to 
 our view. 
 
 As our point of departure we selected, you remember, the 
 familiar fluid, water ; of which we learned the compound nature 
 by an experiment, the simplest, perhaps, and the most striking, 
 in the whole range of chemistry. 
 
 This consisted in dropping on its surface a fragment of the 
 alkali -metal, potassium ; which, at the touch of water, took fire, 
 at the same time liberating from it an inflammable gas called 
 hydrogen. This, upon examination, proved to be the lightest 
 body known ; and we accordingly adopted it as our unitary 
 standard of volume- weight. 
 
 This standard gas we proceeded to liberate, by the same 
 simple means, from two other bodies, less familiar than water, 
 but very well known in the arts and manufactures, viz., muri- 
 atic acid and ammonia, both, in their pure state, gases at 
 ordinary temperature and pressure. 
 
 The further examination of these three sources of hydrogen 
 revealed to us the existence of three gaseous bodies, chlorine, 
 oxygen, and nitrogen, as being respectively associated with 
 hydrogen, in muriatic acid, water, and ammonia. 
 
 The study of chlorine, oxygen, and nitrogen, showed us, in 
 the first, one of the most active of chemical agents ; in the 
 second, the typical supporter of combustion ; in the third, one 
 of the most inert bodies known. 
 
 By the analytic method, applied to decompose the three 
 hydrogen compounds of these gases, and by the synthetic method, 
 employed, so far as available, to reconstruct them, we learned 
 the proportions, as well by volume as by weight, in which chlo- 
 
ESTABLISHMENT OF TYPES. 227 
 
 rine, oxygen, and nitrogen respectively combine with hydrogen, 
 in muriatic acid, water-gas, and ammonia. 
 
 With one unit-volume each of chlorine, oxygen, and nitrogen, 
 weighing respectively 35*5, 16, and 14, we found hydrogen 
 combining in the unit-volume and weight ratios of 1 for the 
 first-named body, 2 for the second, and 3 for the third. 
 
 Notwithstanding this inequality in the number of unit-volumes 
 of the gaseous constituents of these compounds, we found the 
 volumes of the gaseous products to be exactly equal ; measuring, 
 in all three cases alike, 2 unit-volumes. This curious circum- 
 stance proved to us that condensation increases, in these typical 
 cases, pan passu with the number of hydrogen-volumes engaged. 
 
 We thus experimentally established three well-defined models 
 of chemical structure, displayed in the combination of 3 typical 
 elements with the standard element, hydrogen. 
 
 To these types, both of elementary and compound bodies, a 
 fourth in each kind was soon afterwards added ; carbon presenting 
 itself to our notice as the type of the non-volatile elements ; while 
 the richest in hydrogen of its hydrogen-compounds, viz., marsh- 
 gas, contributed its final term to our series of structural models. 
 We had found the three gaseous typical elements successively 
 engaging, within equal product-volumes (double the unit- 
 volume in each case), 1, 2, and 3 volumes of hydrogen ; and 
 now, in the like product -volume of marsh-gas, we found carbon 
 engaging 4 volumes of hydrogen. 
 
 Hence the sort of disjunctive conjunction by which we annexed 
 to our typical series of compounds this singular body, marsh- 
 gas ; alien, with regard to the non- volatile character of its 
 typical constituent, carbon ; cognate, in respect of its ponderal, 
 volumetric, and condensational relations with its standard con- 
 stituent, hydrogen. 
 
 To the establishment of the typical elements, and their typical 
 hydrogen-compounds, succeeded naturally the study of the con- 
 geners in each kind ; and we were thus led to make acquaint- 
 ance with bromine and iodine as analogues of chlorine, and with 
 their respective hydrogen-compounds as analogues of hydro- 
 
228 GENERAL PRINCIPLES AND LAWS. 
 
 chloric acid. With oxygen and its hydrogen-compound, water, 
 we associated, in like manner, sulphur and selenium, and their 
 respective hydrogen-compounds, cast in the structural mould of 
 water-gas. With nitrogen and its hydrogen-compound, ammonia, 
 we connected phosphorus and arsenic, and their ammonia-like 
 combinations with hydrogen. With carbon, lastly, and its 
 hydrogen-compound, marsh-gas, we conjoined, in like manner, 
 silicon, and its hydrogen-compound, formed on the marsh-gas 
 model. 
 
 In these four groups of typical elements and compounds, we 
 recognized the germ of a grand conception that of a natural 
 classification of chemical bodies into genera and species, each 
 distinguished by well-marked characteristics, not excluding in- 
 dividual varieties, but grouping them in subordination to collec- 
 tive laws. 
 
 In the course of these experimental demonstrations, we be- 
 came acquainted with the meaning of the term chemistry., and 
 we obtained our first notions concerning the nature of chemical 
 phenomena. We learned, for example, the characters of ele- 
 mentary as contradistinguished from compound bodies; of 
 chemical combination as contradistinguished from mere me- 
 chanical mixture; of combining proportions, volumetric and 
 ponderal, and of the immutability by which they are charac- 
 terized. 
 
 While thus gradually learning the general principles and laws; 
 of chemistry, we also became familiarized, as we proceeded, with 
 the aspect and uses of chemical apparatus, and with the mani- 
 pulations necessary for their dexterous employment. We gained 
 experience more particularly of the methods in use for gene- 
 rating, collecting, transferring, measuring, desiccating, testing, 
 and weighing gaseous bodies; and for ascertaining the in- 
 fluence of varying temperature and pressure on their bulk and 
 density. 
 
 Our analytic and synthetic operations obliged us to employ, 
 by turns, the powerful agencies of electricity, light, heat, and 
 the specific power we termed chemism, in order to bring about the 
 
SYMBOLIC NOTATION. 229 
 
 desired reactions ; and these we often found to be attended with 
 remarkable physical perturbations ; as, for instance, with the 
 sudden development of light and heat, and often with more or 
 less violent explosion. The means of generating and applying 
 the (so-called) imponderable forces, and of controlling their 
 effects, when excessive or dangerous, were thus brought promi- 
 nently, though cursorily, under our notice. 
 
 Upon the individual characters of the elements and com- 
 pounds thus submitted to investigation we did not dwell at 
 length ; nor, indeed, did we enlarge even upon the general laws 
 with which we met in our course : it was our care to note only 
 such particulars as came within the scope, and promoted the 
 purpose, of our immediate inquiry. 
 
 Doubtless, each subject which we thus touched by the way 
 opened a tempting path to our curiosity ; as the climber, whose 
 appointed aim is the top of the tree, is tempted by the fruit- 
 laden branches he passes in his ascent. 
 
 But though we gladly accepted the incidental information 
 which our experiments naturally threw in our way, we forbore 
 from prolonged digressions, and persisted steadily in the straight 
 course of our inquiry. 
 
 This led us next to the study of the curious and important 
 bodies constituting the nitroxygen series a study in which 
 we broke entirely new ground; quitting the consideration of 
 the typical hydrogen-compounds, each of which only exempli- 
 fies combination in a single fixed ratio, and advancing to the 
 examination of a new and pregnant law of chemistry, that of 
 combination in multiple proportions. 
 
 As our induction thus extended itself, and" our facts began to 
 accumulate, we felt the want of some instrument of record, 
 less periphrastic than ordinary parlance, to epitomise concisely, 
 and to bring graphically and simultaneously under the eye, 
 trains of phenomena which it would else be difficult to grasp 
 and comprehend, firstly, in their mutual relations to each other, 
 and, secondly, in their common dependence on general laws. 
 
 We were thus induced to represent our gas-volumes by squares, 
 
230 PHILOSOPHIC CONCEPTIONS. 
 
 lettered with the initials of the bodies depicted, figured with their 
 relative volume- weights, and forming the germ of a symbolic no- 
 menclature and notation which, in the further stages of our 
 progress, we were enabled at once to enrich and to simplify, by 
 incorporating in it our newly-acquired facts, and eliminating 
 from it forms too cumbersome for practical use, though invalu- 
 able as aids to chemical education. 
 
 That our proportional numbers, abstract at first, might ac- 
 quire a concrete significance, we had to make choice of some 
 system of weight and measure, in terms of whose unitary stan- 
 dards to express those otherwise vague determinations. 
 
 This led us to study the admirable metrical system of the 
 French, which supplied us with our standard of capacity, the 
 litre, and of weight, the gramme ; at the same time teaching us, 
 by means of Greek and Latin prefixes, to express their mul- 
 tiplication and division in decimal progression. 
 
 The weight of the standard volume (1 litre) of our standard 
 element (hydrogen), expressed in terms of our new ponderal 
 unit (1 gramme), gave us the invaluable coefficient O0896 gramme, 
 which, as you remember, we called our "barley-corn-weight" 
 or crith, by means whereof, as a multiplier, we convert the 
 figures representing the mere abstract specific gravities of the 
 various gases and vapours into expressions of their actual or 
 concrete litre- weights. 
 
 The concrete values with which our symbolic expressions 
 thus became clothed increased the power of our system of nota- 
 tion, both as a language for recording phenomena, and as an 
 instrument to assist in their investigation, experimental and 
 theoretic. 
 
 Thus armed, we ventured upon speculative ground; we 
 sought the interpretation of the phenomena we had as yet not 
 observed ; we endeavoured to explain, by a rational hypothesis, 
 the remarkable proportionality, ponderal and volumetric, of 
 chemical reactions; and, with this view, we entered on the 
 inquiry, What is matter ? Of what parts is it composed ? How 
 are these affected by the solid, fluid, and gaseous conditions ? 
 
NATURE OF MATTER. 231 
 
 How are their interspaces filled ? And what, in particular, is 
 the corpuscular construction of a gas ? 
 
 In studying these questions, we were led to admit the three- 
 fold divisibility, molar, molecular, and atomic, of material 
 bodies, and to refrain from asserting their infinite divisibility. 
 The elasticity of gases we attributed to a force connected, in 
 some unknown way, with Jieat ; whose specific relations to dif- 
 ferent bodies, and so-called latency therein, enabled us to under- 
 stand certain, else inexplicable, properties of gases, and to con- 
 ceive these bodies as built up of molecules, or atom-clusters, of 
 which all gases are assumed to contain equal numbers in equal 
 volumes. 
 
 In the light of these conceptions our symbolic language took 
 on a new significance. Besides representing volumes and 
 volume- weights, our squares became pictures for us of molecules 
 and atoms ; whose movements of decomposition and reconstruc- 
 tion we were thus, in imagination, enabled to follow. 
 
 We thus became familiar with the diatomic structure of the 
 typical elementary molecules ; with the monatomic and poly- 
 atomic structure of certain other molecules ; and with the mole- 
 cule-forming and atom-fixing powers of the elementary atoms ; 
 the former of which powers we found to be in the ratio of their 
 atom-weights, while the latter we identified with their atom-freeing 
 and atom-replacing powers, all of which we included in the 
 term quantivalence. These studies led us naturally to touch on 
 and illustrate the principles of quantivalential equipoise, and 
 to embody the idea of quantivalence in our symbolic language, 
 by the addition thereto of appropriate quantivalential coeffi- 
 cients. 
 
 Having learned this much from the study of binary com- 
 pounds, we passed on to consider compounds of a higher order, 
 ternary, quaternary, &c., and the several modes of their genesis 
 from binary compounds ; as, for instance, sometimes by mole- 
 cular, sometimes by atomic inception, each sometimes attended, 
 and sometimes not, by substitutional displacement of atoms from 
 the parent compound. 
 
232 MOLECULES AND ATOMS. 
 
 Examples of ternary compounds generated in each of these 
 modes were supplied to us, in the lecture of to-day, by the four 
 typical groups with whose study we commenced onr course. 
 In our rapid review of these compounds, we noted their usually 
 closely-packed product- volumes, or high vapour-densities ; their 
 frequent tendency to dissociation; their habitual retention of 
 the structural type of their parent compounds ; and the prin- 
 ciples of their progressive or serial development. The last 
 of these interesting compounds was presented to us by marsh- 
 gas; which, in becoming methylic alcohol by inception of 
 oxygen, threw open the gate of a new field of inquiry, and 
 brought our present journey to its term. 
 
 In this rapid retrospective survey, I have not attempted to be 
 encyclopaedic ; my wish has merely been, at parting, to recal the 
 more important of the many deeply-interesting topics which 
 flowed upon us in succession, as out of a living source, from the 
 pregnant conception of the molecular and atomic construction of 
 matter. 
 
 It can now, indeed, be no matter of surprise to you that we de- 
 voted so much time to the consideration of molecules and atoms, 
 and dwelt at so much length upon the methods in use for de- 
 termining the molecular and atomic weights of the elements. 
 These are the foundations of chemical knowledge ; and the table 
 in which they are given, together with their symbols and co- 
 efficients of qiianti valence, though, as we are aware, some of 
 the figures are still doubtful, deserves our close and frequent 
 study. The more of its figures we can bear in mind, the 
 more accurate and ready will our knowledge be, whether for 
 theoretic or practical applications. 
 
 But I must not linger on these themes ; nor, to defer unwel- 
 come separation, trespass still further beyond limits already 
 overpast. 
 
 If, in conclusion, I resume my metaphor, and bid you adieu as 
 fellow-travellers, it is because I deeply feel how much there is in 
 the present transitional condition of chemistry to justify such an 
 expression ; and am almost painfully conscious how narrow is 
 
CONCLUSION. 233 
 
 the explored domain through which the teacher can be your 
 guide, in comparison with those vast regions of truth as yet un- 
 known, in which we are all fellow-students together ! 
 
 And thus, a learner myself, day by day, I can the better appre- 
 ciate your constant and sympathetic attention to my lessons ; 
 and am encouraged the more earnestly to hope that the facts 
 and principles, which, in these few meetings, we have passed in 
 review together, will not merely afford us some present insight 
 into the new doctrines now so deeply, and, let me add, so whole- 
 somely, agitating the chemical world, but will also serve as a 
 firm basis on which, in future conferences, we may build up, 
 stone by stone, the vast superstructure of the chemical edifice. 
 
 THE END. 
 
 LONDON : PRINTED BY WILLIAM CLOWES AND SONS, STAMFORD STRF.KT 
 AVD CHARING CROSS. 
 
RETURN TO the circulation desk of any 
 University of California Library 
 
 or to the 
 
 NORTHERN REGIONAL LIBRARY FACILITY 
 Bldg. 400, Richmond Field Station 
 University of California 
 Richmond, CA 94804-4698 
 
 ALL BOOKS MAY BE RECALLED AFTER 7 DAYS 
 
 2-month loans may be renewed by calling 
 (510)642-6753 
 
 1-year loans may be recharged by bringing 
 books to NRLF 
 
 Renewals and recharges may be made 4 
 days prior to due date. 
 
 DUE AS STAMPED BELOW 
 
 JANQ3ZOOO 
 
 12,000(11/95) 
 
QD31 
 
 H71 
 
 1866 
 
 Hofmann, 
 
 Introduction 
 chemistry-. 
 
 42700 
 to modern