BEYOND THE ATOM 
 
 JOHN COX, M.A. 
 
 Sometime Fellow of Trinity College, 
 Cambridge 
 
 Formerly Professor of Physics in 
 M c Gill University 
 
 Cambridge : 
 at the University Press 
 
 New York: 
 G. P. Putnam's Sons 
 
Cambridge: 
 
 PRINTED BY JOHN CLAY, M.A. 
 AT THE UNIVERSITY PRESS 
 
 With the exception of the coat of arms at 
 the foot, the design on the title page is a 
 reproduction of one used by the earliest known 
 Cambridge printer, John Siberch, 1521 
 
PREFACE 
 
 THIS Essay is an attempt to tell in short compass 
 the romantic story of the discoveries which 
 within the last decade have led us beyond the atom. 
 The great authorities and storehouses of fact are, of 
 course, the books of Rutherford and (recently) Mme 
 Curie on Radioactivity. Excellent statements of the 
 main points are to be found in Strutt's Becquerel 
 Rays, and Soddy's Interpretation of Radium. I 
 have drawn upon all of them; upon the original 
 papers in the current scientific periodicals ; and also 
 on my happy memories of those exciting times when 
 for nine years I was privileged to witness at close 
 quarters and discuss from day to day the work 
 achieved by my friend and colleague Professor 
 Rutherford in the Macdonald Physics Building of 
 McGill University, Montreal, of which I was 
 Director. Nor must* I fail to express my special 
 gratitude to Professor Rutherford for unstinted help 
 and criticism while the work was passing through 
 the press, without which, indeed, it could have had 
 little value as an account of a subject still rapidly 
 growing under his hands. 
 
 J. C. 
 
 February, 1913. 
 
 267395 
 
CONTENTS 
 
 CHAP. PAGE 
 
 I. Introduction The Atom in the Seventies 1 
 
 II. The Vacuum Tube 15 
 
 III. The New Rays 42 
 
 IV. The New Substances 52 
 
 V. Disintegration 67 
 
 VI. A Family Tree 79 
 
 VII. Verifications and Results .... 96 
 
 VIII. The Objective Reality of Molecules . 117 
 IX. The New Atom. . . 131 
 
 Bibliography 147 
 
 Index .... . 149 
 
 PLATE. Tracks of a and /3 particles, facing page 142 
 
CHAPTEK I 
 
 INTRODUCTION 
 THE ATOM IN THE SEVENTIES 
 
 IT is a strange and striking fact that both the 
 beginning and the end of the nineteenth century 
 pre-eminently the century of advance in physical 
 science were marked by extraordinary outbursts 
 of fundamental discoveries. Let us note a few 
 events which may be found grouped within a year 
 or two of the turn of the eighteenth century. 
 
 1798. Rumford and Davy upset the Caloric 
 Theory of Heat, proving that it was a mode of 
 motion affecting the molecules of the hot substance. 
 
 1799 1802. Young upset the Corpuscular 
 Theory of Light, proving that it was a wave-motion 
 in a new medium, the Ether. 
 
 1800. The concept of Energy was first introduced 
 by Young; Volta discovered the Electric Pile; 
 Rumford founded the Royal Institution which, in 
 the opinion of good judges, has done more in the 
 
 c. 1 
 
2 BEYOND THE ATOM [OH. 
 
 century to advance physics than any single University 
 or other body. 
 
 1802. Wollaston first noticed the dark lines in 
 the Solar Spectrum, thus laying the foundation of 
 spectrum analysis. 
 
 1803. Dalton brought forward the Atomic Theory, 
 giving a vast expansion and impetus to Chemistry. 
 
 The turn of the nineteenth century has a group 
 to match. 
 
 1878. Crookes produced the Kathode Rays, 
 claiming that they constituted a new, or fourth, 
 state of matter Radiant Matter. 
 
 1893. Lenard, following up Hertz's discovery that 
 Kathode Rays could pierce a thin sheet of metal, 
 brought them outside the vacuum tube into the air. 
 
 1895. Roentgen, repeating Lenard's work, dis- 
 covered almost by accident the X-rays, and worked 
 out their chief properties. 
 
 1896. Becquerel discovered Radioactivity in 
 uranium. 
 
 / X^1897. J. J. Thomson proved that the Kathode 
 Rays consisted of particles a thousand times smaller 
 than the atom of hydrogen, the smallest atom known. 
 
 1903C Radium was discovered by P. and Mme 
 Curie. 
 
 1896 1906. Rutherford and Soddy established 
 the Disintegration Theory of Radioactivity ; thus 
 founding a new science of Sub-atomic Chemistry. 
 
i] THE ATOM IN THE SEVENTIES 3 
 
 The work of the intervening century brought to 
 light some capital and many minor discoveries. 
 Oerstedt, in 1821, found the first connection between 
 electricity and magnetism in the effect of a current 
 on a neighbouring compass needle; Faraday, in a 
 fortnight of marvellously fruitful work in 1831, laid 
 bare the laws of the induction of currents. Later 
 he found after many searches the first link between 
 electricity, or rather magnetism, and light, and these 
 regions were afterwards brought into one domain by 
 Maxwell's Electro-magnetic Theory of Light which 
 received a splendid verification in the experimental 
 researches of Hertz. 
 
 But mainly the century was occupied with de- 
 veloping and correlating the leading ideas and 
 discoveries which emerged at its birth. To realize 
 the progress made, one must contrast the physical 
 ideas current in 1800 and in 1900. In 1800 Mechanics 
 had already reached a commanding development, 
 and all explanation, even in the more backward 
 sciences, was based on Newton's view of action 
 at a distance as seen in gravitation. Sound and 
 Light were indeed explained mechanically, sound 
 by vibrations of the air, light by the impact of 
 corpuscles. But Heat, Electricity and Magnetism 
 were still isolated regions. Their mysterious pro- 
 perties were attributed to imaginary substances 
 
 12 
 
4 BEYOND THE ATOM [CH. 
 
 Caloric, the Electric Fluid, Magnetic Matter which, 
 because they did not affect the weight of the 
 bodies in which they resided, were known as the 
 Imponderables. 
 
 In 1900 Action at a distance is still employed as 
 a useful figment in mathematical calculations. But 
 the Imponderables are gone. Heat is molecular 
 motion; light an ether vibration; magnetism ether 
 whirls ; electricity ether strain or ether shear. Matter 
 itself may be vortex rings in the ether, or perhaps of 
 electrical origin. Thus Mechanics has invaded the 
 whole field, but it is the mechanics of molecules or of 
 the ether. By extending the laws of motion derived 
 from the study of bodies of sensible magnitude to 
 the molecule and the ether, hypothetical entities 
 for ever far beyond the limits of direct sense percep- 
 tion, a wonderful simplification has been effected. 
 
 The advance has been along two main roads, 
 each lighted by a constellation of seven great 
 names. One of these ways is through the Mechanics 
 of Molecules, covering the theory of Heat and the 
 Gaseous Laws. It was beaten out by Rumford, 
 Mayer, Joule, Helmholtz, Carnot, Kelvin, Clausius. 
 The other track lies through the Mechanics of the 
 Ether, to the modern theories of Light, Electricity 
 and Magnetism. Its pioneers were Young, Fresnel, 
 Ampere, Faraday, Kelvin, Maxwell and Hertz. And 
 
i] THE ATOM IN THE SEVENTIES 5 
 
 midway in the map of the century stand the two 
 great landmarks : 
 
 Conservation of Energy ... 1847 
 Origin of Species ... 1859. 
 
 Insight into nature has grown by a process of 
 deepening as well as widening, but above all of unifica- 
 tion. It has been like the clearing of the mist to the 
 watcher on a mountain top. Here and there the sun 
 strikes a finger through it and touches into dazzling 
 light the spires of a city, the silver bend of a river, or 
 a blue gleam of the distant sea, set each in its frame 
 of white vapour; and then at the sudden magic of 
 some wandering air the curtain rolls off and leaves 
 the landscape not only clearer where it could be seen 
 before, but a rounded whole, right out to the circling 
 horizon, the bearings of its separate details and their 
 linked connections of road and river and railway 
 spread like a map to the view. Here and there the 
 curling wisps of cloud may still conceal a hollow in 
 the hills or a sleeping lake ; and always there is the 
 baffling horizon, broadening the higher we rise, and 
 impervious as ever to our gaze. 
 
 With this preliminary survey of the general course 
 of physics in the nineteenth century, we shall now 
 turn to the discoveries of the last decade, which are 
 already seen to bear chiefly on the atomic structure 
 of matter, and trace in outline the steps by which 
 
6 BEYOND THE ATOM - [OH. 
 
 physicists have been led to advance beyond Dalton's 
 Atomic Theory and proceed beyond the atom. 
 
 Atomic theories date back at least to Democritus. 
 Dalton's service lay in this, that he gave precision to 
 the idea for the purposes of modern chemistry. When 
 a chemical compound is formed the substance con- 
 cerned will not unite in haphazard amounts, as in a 
 mere mixture, but only combine indefinite proportions. 
 Thus nitrogen forms five different compounds with 
 oxygen, in which 28 parts by weight of nitrogen are 
 joined respectively with 16, 32, 48, 64 and 80 parts 
 by weight of oxygen. The oxygen in the different 
 compounds is for a given amount of nitrogen in the 
 proportion of 1, 2, 3, 4, 5. Hence Dalton's view. 
 The smallest amount of one of these compounds 
 which could exist while still retaining the properties 
 of the compound was called a molecule of that com- 
 pound. The molecule was composed of atoms of 
 nitrogen and oxygen. The atoms of nitrogen were 
 all identical and had the same weight ; and so for 
 oxygen and all the other elements, or simple sub- 
 stances. The atom of hydrogen, the lightest, being 
 taken as a standard, that of nitrogen weighed 14 and 
 oxygen 16 times as much. For good reasons the 
 compounds of nitrogen and oxygen were held to 
 result from the union of two atoms of nitrogen (each 
 14) with 1, 2, 3, 4 or 5 atoms of oxygen (each 16). 
 The law of definite or multiple proportions, as found 
 
i] THE ATOM IN THE SEVENTIES 7 
 
 by experiment, was accounted for, and the chemist 
 had a convenient means of describing, systematizing, 
 and thinking about the complicated facts he had to 
 deal with ; which, indeed, is all that can be asked of 
 a theory or working hypothesis. For, as Sir Joseph 
 Thomson says, 'a theory is a policy rather than a 
 creed/ For the purposes of the chemist the atomic 
 theory is as serviceable to-day as it has been through- 
 out the century, in spite of the modifications which 
 the work of the last ten years has compelled in our 
 conception of the atom. 
 
 Meanwhile the physicists employed the molecule 
 (or in the case of elements the atom) to elucidate 
 those properties of matter which fall short of chemical 
 change. The facts of cohesion indicate that molecules 
 can exert enormous forces of attraction on each other, 
 but only when brought within inconceivably minute 
 distances. Thus a broken solid cannot be mended by 
 forcing the parts together by hydraulic pressure, 
 however great. The molecules cannot be brought 
 again within the range of their mutual attractions. 
 To effect this fusion must be resorted to, or some 
 cement which is able to come into such intimate 
 contact at ordinary temperatures. On the other 
 hand, the molecules are supposed to be in rapid 
 motion, and the states of matter, as we know it by 
 the senses, solid, liquid and gaseous, result from a 
 balance between the attractive forces and the motions. 
 
8 BEYOND THE ATOM [OH. 
 
 In solids the molecules, or at least most of them, 
 are permanently anchored in fixed positions, locked 
 together by their mutual forces; and yet are in 
 perpetual agitation, either of vibration or rotation, 
 about those positions. It may seem outrageous to 
 claim that a block of gold or of lead, apparently so 
 quiescent, is really the theatre of violent tremors 
 and motions among its myriads of particles. But 
 the senses are not safe guides. The eye cannot 
 discriminate the spokes of a wheel in comparatively 
 sluggish motion. And there is plenty of positive 
 evidence that the molecules do travel, even in solids. 
 If the block of gold be left in contact with the lead 
 and examined from time to time, traces of each metal 
 will presently be found throughout the other. Again, 
 if, as Rumford showed, heat is motion, it must be 
 molecular motion. The expansion of solids no less 
 than liquids with rise of temperature is consistently 
 explained if temperature corresponds to the average 
 molecular energy of motion. 
 
 In liquids, though the bonds are partially relaxed 
 and the molecules may slip freely over each other 
 from one point to another, they are never far enough 
 apart to escape the constraint of their mutual forces. 
 The facts of evaporation and diffusion, much more 
 marked than in solids, compel a belief in their mo- 
 bility. 
 
 In gases at last the molecules are for most of their 
 
i] THE ATOM IN THE SEVENTIES 9 
 
 time free from mutual constraint. Since a cubic inch 
 of water occupies 1600 cubic inches when converted 
 into steam, the spaces between the molecules must be 
 very large compared with the size of a molecule. 
 And since a vapour expands instantly to occupy any 
 new space opened to it, the velocity of travel must be 
 great. In fact a gas is conceived as a crowd of par- 
 ticles travelling at high speed for the most part in 
 'free paths,' perpetually interrupted by collisions or 
 encounters with each other and with the sides of the 
 containing vessel. The pressure on the vessel is the 
 averaged result of the blows from the molecules which 
 bombard it ; just as a high-pressure jet of water may 
 be used in mining to cut away the side of a hill. 
 
 On this theory it has been possible to find consis- 
 tent explanations, not only of the ordinary properties 
 of matter, but of the main laws of gaseous pressure. 
 These are (1) the law of Boyle, that the pressure of a 
 gas is proportional to its density, and (2) the law of 
 Charles and Gay-Lussac, that the pressure is pro- 
 portional to the temperature of the gas measured 
 from absolute zero. 
 
 More interesting from the point of view of this 
 essay are the inferences which, conversely, have been 
 drawn from known physical facts with regard to 
 the molecules themselves. Different investigators, 
 Loschmidt, Maxwell, Stefan, Stoney, Kelvin starting 
 from such apparently unconnected facts as diffusion, 
 
10 BEYOND THE ATOM [CH. 
 
 viscosity and conduction of heat in gases, the thick- 
 ness and surface tension of a soap bubble just before 
 it breaks, and the electric action between thin layers 
 of zinc and copper, have nevertheless arrived at 
 results which are in fair agreement at least so far as 
 the order of the magnitudes concerned. The numbers 
 generally accepted for hydrogen and oxygen, for 
 instance, are given by Maxwell as follows : 
 
 Hydrogen Oxygen 
 
 Mass of molecule (hydrogen =1) ... 1 16 
 Velocity (of mean square) in metres 
 
 per second at C 1859 fl - 465 
 
 Mean free path in millimetres ... 965 x 10~ 7 560 x 10~ 7 
 
 Collisions (millions per second) ... 17750 7646 
 
 Diameter (millimetres) 5'8xlO~ 7 7'6xlO~ 7 
 
 Mass (in grams) 46 x lO" 25 736xlO~ 25 
 
 About two million molecules of hydrogen in a row 
 would occupy a millimetre, i.e. the twenty-fifth of an 
 inch. Fifteen thousand million million million of 
 them would weigh a grain. When the gas is at 
 freezing-point each molecule makes on the average 
 17,750 million collisions per second at each of which 
 its speed is changed and diverted. Yet on the average 
 it has a speed of over six thousand feet per second, 
 moving about a quarter of a millionth of an inch 
 between collisions. 
 
 The number of molecules in a cubic centimetre of 
 hydrogen is 3'8 x 10 19 , or about six hundred million 
 
i] THE ATOM IN THE SEVENTIES 11 
 
 million million per cubic inch. And this number 
 holds for all gases, since one of the consequences of 
 the kinetic theory of gases is Avogadro's law, that the 
 number of molecules in a cubic centimetre at a given 
 temperature and pressure is the same for every gas. 
 
 As to the structure of the atoms, the most tempt- 
 ing view in the seventies was the vortex-ring theory 
 of Sir William Thomson, based on Helmholtz's in- 
 vestigation of vortex motion in a perfect fluid. Every- 
 one knows the rings which can be blown by a skilful 
 smoker. Even in a viscous fluid like air they show 
 considerable permanence. But Helmholtz had proved 
 that in a frictionless fluid, such rings can neither be 
 created nor, if once in existence, destroyed. Each 
 ring always consists of the same portion of the fluid, 
 and its ' strength,' i.e. the product of its cross section 
 by its velocity of rotation, remains the same. No 
 ring can cut across another, or across any of its own 
 convolutions. Hence if two rings are linked, they can 
 never be separated, or if a ring is knotted on itself, it 
 can never be untied. Moreover, two rings at an en- 
 counter exhibit attraction or repulsion according to 
 their directions of rotation. 
 
 In these properties Thomson recognized many of 
 the necessary characteristics of an atom. 'In the 
 first place,' says Maxwell, ' the vortex ring is quanti- 
 tatively permanent as regards its volume and its 
 strength, two independent quantities. It is also 
 
12 BEYOND THE ATOM [CH. 
 
 qualitatively permanent as regards its degree of im- 
 plication, whether " knottedness " on itself or " linked- 
 ness " with other vortex rings. At the same time it 
 is capable of infinite changes of form, and may execute 
 vibrations of different periods, as we know that mole- 
 cules do. And the number of essentially different 
 implications of vortex rings may be very great with- 
 out supposing the degree of implication of any of them 
 very high.' 
 
 There is a fascination about the vortex-ring 
 theory because of the prospect of simplification it opens 
 out. It unites the two main lines of research we have 
 traced through the century, since the structure of 
 matter is reduced to the properties of motion in a 
 frictionless fluid which may be taken to be the ether. 
 The atom is no longer a foreign body embedded in an 
 alien medium, yet able to communicate vibrations to 
 it ; but is a portion of the universal ether itself, only 
 differentiated, and yet permanently differentiated, 
 from the rest by its mode of motion. 
 
 The position reached in the seventies may be 
 summed up in a famous passage from Maxwell's 
 article on the atom in the Encyclopaedia Brit. 9th 
 Ed. 1875. Having discussed the volume, mass and 
 other characteristics of the molecule, or atom, in- 
 cluding its rates of vibration as revealed by the 
 spectroscope, and found them identical for all 
 molecules of the same kind, he proceeds: 
 
i] THE ATOM IN THE SEVENTIES 13 
 
 'It is the equality of these space- and time- 
 constants for all molecules of the same kind which 
 we have next to consider. We have seen that the 
 very different circumstances in which different mole- 
 cules of the same kind have been placed have not, 
 even in the course of many ages, produced any 
 appreciable difference in the values of these constants. 
 If, then, the various processes of nature to which 
 these molecules have been subjected since the world 
 began have not been able in all that time to produce 
 any appreciable difference between the constants of 
 one molecule and those of another, we are forced to 
 conclude that it is not to the operation of any of 
 these processes that the uniformity of the constants 
 is due. 
 
 ' The formation of a molecule is therefore an event 
 not belonging to that order of nature under which we 
 live. It is an operation of a kind which is not, so far 
 as we are aware, going on on earth or in the sun or 
 the stars, either now or since these bodies began to 
 be formed. It must be referred to the epoch, not of 
 the formation of the earth or of the solar system, but 
 of the establishment of the existing order of nature, 
 and till not only these worlds and systems, but the 
 very order of nature itself is dissolved, we have no 
 reason to expect the occurrence of any operation of 
 a similar kind. 
 
 'In the present state of science, therefore, we have 
 
14 BEYOND THE ATOM [CH. 
 
 strong reason for believing that in a molecule, or if 
 not in a molecule, in one of its component atoms, we 
 have something which has existed either from eternity 
 or at least from times anterior to the existing order 
 of nature. But besides this atom, there are immense 
 numbers of other atoms of the same kind, and the 
 constants of each of these atoms are incapable of 
 adjustment by any process now in action. Each is 
 physically independent of all the others/ 
 
 When this passage was written, the march of 
 science had apparently reached a pause. During the 
 century the whole field of natural knowledge had 
 been for the first time systematically ransacked. The 
 first rich harvest of great discoveries had been made ; 
 and here, with an alluring speculation it is true, the 
 path of theory seemed to end in a barrier to further 
 progress. It was excusable that many should feel 
 there was little left for the future except to glean 
 here and there an ear which had been overlooked or 
 to conquer perhaps another decimal in the accuracy 
 with which a physical constant was known. But in 
 science less than anywhere is it safe to venture a 
 negative, and the world was on the verge of discoveries 
 which have given a Sophoclean irony to Maxwell's 
 words. 
 
ii] THE VACUUM TUBE 15 
 
 CHAPTER II 
 
 THE VACUUM TUBE 
 
 EVER since the invention of the Induction Coil 
 afforded the means of passing high-tension currents 
 through partially exhausted tubes, anyone watching 
 the strange and beautiful phenomena within the tube 
 must have felt instinctively that here was the border- 
 land of discovery, here if anywhere might be sought 
 further light on the ultimate nature of matter and of 
 electricity, and their mutual relation. 
 
 The terminals of the coil are joined up to elec- 
 trodes of platinum wire fused through the glass. In 
 Faraday's nomenclature the electrode by which the 
 current enters, connected therefore to the positive 
 terminal, is called the Anode : the negative electrode 
 is called the Kathode. With a long tube of air at 
 ordinary pressure no current will pass. But on partial 
 exhaustion the current leaves the spark gap of 
 the coil and takes to the tube, at first in the form of 
 a narrow stream of crimson light. When the pressure 
 is reduced to two or three millimetres of mercury, 
 about one two-hundredth of an atmosphere, the 
 Geissler vacuum is reached. The kathode is now 
 seen to be encased in a thin dark space, outside 
 
16 BEYOND THE ATOM [CH. 
 
 which is a glow of soft violet light. Then follows 
 another dark space, and beyond it the tube is filled 
 as far as the anode with flickering sheets of light 
 the Striae separated by darker spaces. The anode 
 itself is tipped with a vivid speck of light. The colour 
 of the striae depends on the gas which last filled the 
 tube. For hydrogen it is reddish, for nitrogen a 
 deeper red, for carbonic acid a pale green. 
 
 It was Sir William Crookes who, in his celebrated 
 lecture to the British Association in 1878, first brought 
 prominently forward the results of carrying the vacuum 
 much farther. As the pressure decreases the dark 
 space about the kathode expands, driving the striae 
 before it, until at a pressure of about a millionth of 
 an atmosphere the dark space reaches the walls of 
 the tube. The interior of the tube is then dark 
 except for a faint cone of violet light which seems to 
 shoot out from the kathode in straight lines ; but the 
 walls of the tube shine with a brilliant fluorescent 
 light, yellowish green for ordinary glass, steel blue for 
 flint glass. 
 
 Crookes showed a number of very beautiful tubes 
 constructed to prove the following points : 
 
 (1) The faint violet light shoots out from the 
 kathode at right angles to its surface, and proceeds 
 in straight lines, like light, independently of the posi- 
 tion of the anode ; whereas at the lower vacuum the 
 column of the striae bends round corners, even into 
 
n] THE VACUUM TUBE 17 
 
 side-tubes, till it reaches the anode. Hence he gave 
 this light the name of Kathode Rays. 
 
 (2) Where the rays strike they cause fluorescence, 
 as on the walls of the tube. If a maltese cross of 
 metal be set up within the tube in the path of the 
 rays, it casts a shadow on the far end. Objects, such 
 as diamonds, rubies, shells, minerals, exposed to the 
 rays within the tube shine with brilliant and charac- 
 teristic colours. 
 
 (3) The rays exert mechanical pressure. If 
 directed on to light vanes attached to an axle, they 
 can be made to turn little mills, or drive a wheel along 
 glass rails, as in the ' railway tube.' 
 
 (4) They heat the object struck. By using a 
 kathode shaped like a small concave mirror they may 
 be converged on to a piece of platinum, which soon 
 becomes red hot, and may even be fused. 
 
 (5) When a magnet is brought near the tube, the 
 stream of rays is deflected, moving across the lines of 
 force of the magnetic field in the same way as a 
 negative current would do. 
 
 Crookes concluded that the kathode rays were 
 streams of minute particles travelling at very high 
 speed and each conveying a negative charge. If this 
 were so, then two streams travelling side by side 
 should diverge, owing to the mutual repulsion of their 
 charges. In another tube this was shown to be the 
 case. Crookes therefore held that the kathode rays 
 
 c. 2 
 
18 BEYOND THE ATOM [OH. 
 
 constituted a fourth state of matter, neither solid, 
 liquid, nor gaseous, and called it Radiant Matter. 
 
 For nearly twenty years a controversy raged as to 
 the nature of kathode rays. For the most part the 
 English school held with Crookes that they were 
 streams of particles ; with some difference of opinion 
 whether they were fragments of the metal electrodes 
 torn off by the violent electric forces, or, perhaps, 
 the remainder molecules of the gas, now at liberty in 
 the high vacuum to move with greater ease over 
 longer free paths under the influence of the electric 
 field. 
 
 In Germany it was maintained that they were not 
 particles at all, but an ether phenomenon, of the 
 nature of ultra-violet light. Perrin, however, proved 
 that they carried a negative charge by receiving them 
 in a small metal cylinder fixed within the tube and 
 connected by a wire fused through the glass with an 
 external electroscope. When the rays fell into the 
 cylinder the electroscope showed a growing negative 
 charge ; but when they were deflected by a magnet 
 so as to miss the cylinder, the action ceased. In 1893 
 Lenard, following up an observation of Hertz that the 
 rays could penetrate a thin sheet of metal, succeeded 
 in bringing them outside the tube. A small window 
 in the end of the tube was sealed with a thin foil of 
 aluminium. It was only one thirty-fifth of an inch in 
 diameter, so that the foil might stand the atmospheric 
 
n] THE VACUUM TUBE 19 
 
 pressure. When the rays were directed on to the 
 window, they issued through the foil into the open air. 
 
 Lenard's experiment certainly favoured the 
 English view, since the foil used was impervious to 
 any form of light. But it was J. J. Thomson who in 
 1897 finally cleared up the nature of the kathode 
 ray. His experiment, as ingenious as it was difficult, 
 deserves special attention, since in principle it has 
 been applied to test the nature of all the rays to be 
 dealt with subsequently. The principle is based on 
 Crookes' magnetic deflection experiment. 
 
 A particle carrying an electric charge projected 
 across a uniform magnetic field at right angles to 
 the magnetic lines of force is by known electric laws 
 subject to a deflecting force perpendicular at each 
 point to the direction in which the particle is moving. 
 But this is a well known case of motion. A stone 
 whirled round at the end of a string is subject to 
 just such a force, since the string is always pulling it 
 along the radius. The electric particle will therefore 
 describe a circle like the stone. In fact with proper 
 arrangements the kathode rays can be curled up 
 into a circle within the tube, and the radius of the 
 circle can be measured, or if the circle is not com- 
 pleted, its radius can be calculated from the observed 
 curvature of the path. 
 
 Suppose that the kathode ray consists of particles 
 each of mass m, carrying a charge e, at a speed v. 
 
 22 
 
20 BEYOND THE ATOM [OH. 
 
 Then by exposing them to a known magnetic force, 
 and observing the radius of the circle described, it 
 should be possible to relate the quantities e, m, v. 
 In fact, if H is the strength of the magnetic field 
 and r the radius of the circle, we know from the 
 laws of electricity that the deflecting force is H . e. v; 
 and this must be the force required to make the 
 particle go round a circle of radius r, which from 
 
 elementary mechanics is known to be . 
 Thus 
 
 e v 
 
 and ~ = w- 
 
 m Hr 
 
 Since H and r are known numbers, we have a 
 
 & 
 relation between the ratio, , of the charge carried 
 
 by a particle to its mass, and the velocity v with 
 which the particle is moving. 
 
 To determine each of these quantities separately 
 a second equation must be found. This has been 
 done in several ways. One method employed by 
 J. J. Thomson is the following, in which the magnetic 
 deviation of the rays is compared with that produced 
 by an electric field. The kathode rays from C are 
 reduced by the diaphragms AB to a narrow pencil 
 which passes between two electrified plates D, E and 
 falls on a phosphorescent screen PP'. The negative 
 
II] 
 
 THE VACUUM TUBE 
 
 21 
 
 particles of the ray are repelled from the negative 
 plate and attracted towards the positive by a force 
 Xe, if X is the electric force between the plates. 
 
 A magnetic field (not indicated in the figure) is 
 now applied, as in the previous experiment, in such 
 a direction as to oppose the electric force, and its 
 strength is varied till the spot of light P' is brought 
 
 Fig. 1. 
 
 back to its undeviated position P. When this balance 
 is effected, 
 
 Hev = Xe, 
 
 or 
 
 X 
 H 
 
 and when v is known, -- follows from the previous 
 
 equation. 
 
 The results of this brilliant piece of work, confirmed 
 in essentials by many subsequent investigations, were 
 indeed surprising. 
 
 It appeared that the velocity of the kathode 
 particles might range from one-thirtieth to one-thir^ 
 
22 BEYOND THE ATOM [OH. 
 
 the speed of light, i.e. from about 6000 to 60,000 
 miles per second according to the electric pressure 
 applied to the tube, or the pressure and nature of 
 the residual gas within it. These were astonishing 
 speed limits. But even more significant was the fact 
 
 ^> 
 
 that the ratio of the charge carried to the mass of 
 m 
 
 the carrier particle, came out the same, whatsoever 
 the nature or pressure of the gas in the tube, or the 
 metal used as kathode. 
 
 While this result practically established the view 
 that the kathode ray consisted of charged particles, 
 it could no longer be supposed that the particles were 
 
 either molecules of the gas or fragments torn off the 
 
 & 
 
 kathode. The actual value of was about ten 
 
 m 
 
 million (10 7 ) if e is measured on the electro-magnetic 
 system of units. 
 
 Now there is a well known case in which 'ions' or 
 travelling particles carry electric charges. In the 
 electrolysis of water by the electric current the atoms 
 of hydrogen and oxygen moving in opposite directions 
 convey known quantities of electricity per gram of 
 hydrogen or oxygen liberated. It is easy by com- 
 paring these quantities to show that for the atom of 
 
 p 
 
 hydrogen, acting as carrier, is about ten thousand 
 
 m 
 
 (10 4 ). The ratio for the kathode particle, ten million, 
 
n] THE VACUUM TUBE 23 
 
 is thus about a thousand times greater than for the 
 hydrogen atom in electrolysis. 
 
 There are only two possible explanations, or 
 perhaps a third may be found in the combination 
 of the two, though this is unlikely. Either the 
 charge of the kathode particle is a thousand times 
 greater than that of the hydrogen atom, the masses 
 being the same ; or, if the charges are the same, the 
 mass of the kathode particle must be a thousand 
 times less than that of the hydrogen atom, the smallest 
 body hitherto known to science ! 
 
 Before accepting either of these alternatives it is 
 well to see if the strange facts of the kathode ray 
 can be supported by other instances. This is the 
 case. When we come to the study of radioactivity 
 we shall make acquaintance again and again with 
 charged particles for which this same high value of 
 
 the ratio -'- holds good. At the present stage we 
 
 shall only adduce one singularly interesting piece of 
 evidence which comes from quite a different quarter. 
 In the last years of Faraday's life, when his memory 
 was gone and his faculties were clouded, the arrival 
 of a new Steinheil spectroscope at the Royal Institution 
 brought him up from his retirement at Hampton 
 Court to try for one more discovery. He set the 
 source of light in a powerful magnetic field and 
 examined the spectrum expecting some change in 
 
24 BEYOND THE ATOM [CH. 
 
 the lines. The experiment was a failure. But we 
 now know that his wonderful instinct had guided him 
 to try all the right arrangements. Nothing was 
 wanting for success except more power in the 
 spectroscope, and thirty years later, in 1897, Zeeman 
 applying the Rowland diffraction grating found the 
 effect which Faraday looked for in vain. Certain 
 lines are displaced, doubled, trebled, and in some 
 cases split into six, and variously polarized. The 
 Zeeman effect could be explained on Lorenz's theory 
 of radiation, that the light vibrations were started by 
 charged particles either moving in orbits or vibra- 
 ting within the atom ; and the resulting formula for 
 the displacement of a line showed that it depended 
 
 on the strength of the magnetic field and the ratio - 
 
 for the vibrating particle. A comparison of the 
 formula with the observed displacements showed 
 that the charge e must be negative and that the 
 
 S) 
 
 ratio for the light vibrators was again about 10 7 , the 
 
 same as for the kathode particles. 
 
 It seems therefore impossible to doubt that there 
 
 a 
 
 exist particles for which the ratio is 10 7 , or a thou- 
 
 m 
 
 sand times greater than for the hydrogen atom in 
 electrolysis. 
 
 Returning to our alternatives, are we to believe 
 
n] THE VACUUM TUBE 25 
 
 that this is because the charge carried is a thousand 
 times as great ? Or that the mass of the carrier is a 
 thousand times less than t an atom of hydrogen? 
 There is much presumptive evidence against the 
 former view, evidence tending to show that electricity, 
 like matter, is atomic in structure, i.e. that there is 
 a definite minimum or unit charge of electricity, and 
 that all other charges are multiples of this elementary 
 ' atomic * or ' ionic ' charge. Thus in electrolysis, when 
 several cells containing different substances are joined 
 up in series, so that the same current passes through 
 each, through water in one, hydrochloric acid in 
 another, copper sulphate in a third, silver chloride 
 in a fourth, whatever the substance under decom- 
 position, each 'ion,' whether atom or molecule, is 
 found to carry the same ' ionic ' charge. There are 
 other ways of producing l ions,' or charged particles. 
 They are given off from the hot filament of an incan- 
 descent lamp, and from zinc exposed to ultra-violet 
 light. As we shall see, they are caused by the passage of 
 X-rays through a gas, and by the radiations from radio- 
 active substances. J. J. Thomson has succeeded in 
 measuring the charge associated with them in all these 
 instances, and in every case it turns out to be this fun- 
 damental ' ionic ' charge or some simple multiple of it 
 The first case to be successfully attacked was that 
 of the ions produced by X-rays. The charge carried 
 by them was measured by Thomson in 1898, by 
 
26 BEYOND THE ATOM [CH. 
 
 means of one of the most beautiful experiments ever 
 devised, an experiment based on C. T. R. Wilson's 
 work on the formation of clouds. Wilson had studied 
 the conditions under which water vapour condenses 
 into drops upon sudden expansion. A vessel filled with 
 air standing over water, and therefore saturated with 
 water vapour was arranged so that the space could be 
 suddenly expanded by a measurable fraction of its 
 volume. At first a very small expansion formed a cloud, 
 for the vapour condensed upon the dust particles 
 present. After several clouds had been formed and 
 allowed to settle, so that the dust was all carried 
 down into the water, no further condensation could 
 be obtained in the dust-free air unless the space was 
 expanded by at least a quarter of its volume. Then 
 a few drops appeared, until the expansion reached 
 rather more than one-third from 1 to 1*38 when a 
 dense cloud again appeared. Now it is possible to pro- 
 duce within the vessel 'ions,' i.e. small particles with an 
 electric charge, such as are in question in the kathode 
 ray and kindred phenomena, and Wilson had shown 
 that these can act as nuclei for condensation as well as 
 the dust particles. If they are present, then a cloud 
 appears whenever the sudden expansion reaches the 
 value l'2o, i.e. the original space is increased by a 
 quarter of its volume. Thomson noticed that the 
 cloud greatly increased in density when the expansion 
 reached 1*31, and suggested that this was the point 
 
n] THE VACUUM TUBE 27 
 
 when the positive ions came into action, the negative 
 alone proving effective at the lower value T25. In 
 another experiment Wilson succeeded in separating 
 the negative and positive ions in the same vessel and 
 proved that this was the case. 
 
 The next step was to count the number of droplets 
 formed. This was effected by timing the rate at 
 which the cloud in the vessel slowly settled down. 
 For Stokes had given a formula for the rate of fall of 
 a small sphere in a viscous gas like air, a formula 
 which depended on its radius. From the observed rate 
 of fall the radius of the droplets could thus be found. 
 And since the total amount of water precipitated by 
 a given expansion could easily be calculated, it was 
 only necessary to divide this amount by the volume 
 of a droplet, and the number of the droplets was found. 
 
 We are now prepared to understand how Thomson 
 measured the charge on an ion. Two horizontal 
 plates were fitted in Wilson's expansion vessel, and 
 connected in circuit with a weak battery and a current- 
 measurer. Ions were produced between the plates, 
 and upon expansion the cloud was formed. Assuming 
 that each droplet represented an ion, the ions could be 
 counted. As they travelled to the two plates, positive 
 and negative each seeking the plate of opposite kind, 
 they carried a current round the circuit, and from its 
 intensity, now that the number of ions concerned was 
 known, the charge borne by each could at last be 
 
28 BEYOND THE ATOM [OH. 
 
 determined. The value deduced by Thomson was 
 e = 3'4 x 10~ 10 electrostatic units, 
 
 approximately the same that is carried by the hydro- 
 gen atom in the electrolysis of water. 
 
 There were obvious difficulties and sources of 
 error in Thomson's experiment which rendered his 
 result liable to a variation of as much as thirty per 
 cent. And it involved one assumption that required 
 justification, viz. that all the ions were employed as 
 nuclei and that no droplet contained more than one 
 ion. This point was cleared up by an ingenious 
 variation of the experiment due to H. A. Wilson, 
 who compared the rate of fall of the cloud under 
 gravity with its velocity under the influence of an 
 electric field. When the electric field was applied 
 the cloud at once separated into two or sometimes 
 three clouds with different speeds corresponding to a 
 single, double or triple charge. Some of the droplets 
 can therefore absorb two or even three ions. Confining 
 his observations to the cloud which carried the 
 smallest charge Wilson deduced results which, though 
 still subject to variations of as much as thirty per 
 cent., gave a mean value for the ionic charge of 
 e = 3'2xlO~ 10 which was only slightly lower than 
 Thomson's value 3'4 x 10- 10 . 
 
 According to the most recent researches, based 
 on radioactive phenomena and also on the theory o 
 
n] THE VACUUM TUBE 29 
 
 radiation, this value is in fact about thirty per cent, 
 too low, and Rutherford has made an interesting 
 suggestion to account for the error. He attributes it 
 to the fact that no allowance was made for evaporation 
 from the drops during the fall of the cloud, so that 
 their radius, as determined by Stokes' formula, would 
 come out too small, their number consequently too 
 great, and the charge carried by each too small again. 
 To such minute points must the mind be alert in 
 these difficult investigations. Patience and care 
 will in the end attain accuracy in estimating the 
 effect of any one factor ; the task of genius is to divine 
 what factors should be included, to watch that none 
 be overlooked. 
 
 To obviate this error Millikan and Regener have 
 repeated the experiment, using droplets of oil which 
 are not subject to evaporation in the same way. 
 Their results gave e 4'06 x 10~ 10 , in close agreement 
 with the most recent values obtained otherwise. 
 
 The same charge has been found in association 
 with minute solid particles, such as those which 
 constitute a cloud of tobacco smoke. The smoke is 
 blown into a glass vessel containing ionized air and 
 powerfully illuminated from the side. Viewed under 
 a microscope the particles appear as brilliant silver 
 points, so that it is no longer necessary to treat the 
 cloud as a whole but the actual displacements of 
 individual particles can be followed. When an 
 
30 BEYOND THE ATOM [OH. 
 
 electric field is applied, they separate instantly into 
 three groups, one travelling with the field, therefore 
 positively charged ; one against it, negatively charged; 
 and one which remains in place and is therefore 
 neutral, consisting of those particles to which an 
 equal number of ions of opposite signs have attached 
 themselves. In this way Ehrenhaft has found the 
 ionic charge to be 4'6 x 10~ 10 ; de Broglie 4'5 x 10~ 10 . 
 
 But a method devised by Townsend in 1900 
 admits of far greater accuracy. Without measuring 
 the actual charge at all, he showed by comparing the 
 coefficient of diffusion of ions with their 'mobility,' 
 i.e. their velocity under the action of an electric field, 
 that for all types of ions examined the charge carried 
 must be the same as that carried by the atoms in 
 electrolysis, with a possible error of less than one 
 per cent. 
 
 The cumulative effect of all these instances, where 
 the ions are produced in different gases, under 
 different circumstances, by different means, but always 
 are found to carry the same charge, forces the con- 
 viction that there exists an ultimate charge or 'atom' 
 of electricity ; that the particles of the kathode ray 
 can be no exception ; and that this atomic charge, or 
 some very simple multiple of it, is the charge they 
 carry, not a charge more than a thousand times as 
 great. But if this be so then we are forced back on 
 the other alternative, and the conclusion is irresistible 
 
ii] THE VACUUM TUBE 31 
 
 that in the kathode ray we are dealing with an 
 entity a thousand times less than anything previously 
 conceived by science. And since these particles are 
 identical, whatever the gas from which they are 
 derived, apparently they are a constituent of all 
 matter. A modification of Prout's hypothesis is 
 thus suggested. Front held that the atoms of all 
 other elements are built up of hydrogen atoms. It 
 now appears that all atoms, that of hydrogen included, 
 are at least partly constituted of particles of the same 
 kind. 
 
 It is time to give a name to these new bodies. At 
 first they were called corpuscles, but they are now 
 known as Electrons. An electron, then, is a minute 
 particle with an apparent mass about one- thousandth 
 that of a hydrogen atom, and bearing a negative charge 
 
 Canal Rays. 
 
 Meanwhile (1896) Goldstein observed a new type 
 of ray within the vacuum tube. Employing a per- 
 forated kathode, he noticed cones of violet light pro- 
 jected from each of the holes behind the kathode. 
 These rays he called ' Canal rays ' from the passages 
 in the kathode through which they issued. They were 
 studied by Wien, by methods entirely similar to those 
 just described. 
 
32 BEYOND THE ATOM [OH. 
 
 Wien found that they could be deflected by 
 magnetic and electric fields though much more feebly 
 than the kathode rays, and the deflection was in the 
 
 /> 
 
 apposite direction. The value of was variable but 
 
 never greater than 10 4 . 
 
 The canal rays must therefore consist of particles 
 of atomic size and bearing a positive charge. 
 
 After a long interval their true nature has just 
 been elucidated by some remarkable researches of 
 J. J. Thomson (1910-12). He employs a kathode 
 with a single long and very narrow perforation, 
 leading to a large discharge tube. The particles 
 projected through this passage are subjected to the 
 action of an electric and a magnetic field mutually 
 at right angles and both perpendicular to the path 
 of the particles. Theoretically, the electric displace- 
 ment should be proportional to the square of the 
 magnetic displacement, and thus when the particles 
 strike a screen placed to receive them, they should 
 be found on the arc of a parabola. Thomson places 
 a photographic plate within the discharge tube, and 
 the plates, on developing, show in fact a number of 
 such parabolic arcs, corresponding to as many different 
 types of projected particles. The curves are measured 
 and the latus rectum of each parabola gives the value 
 
 of for the particles which formed it. 
 m 
 
n] THE VACUUM TUBE 33 
 
 A discussion of the results reveals the presence of 
 many types. Thus in an oxygen tube were found 
 (1) ordinary molecules of oxygen each consisting of 
 two atoms united ; neutral atoms ; atoms with single 
 and double positive charges ; atoms with one negative 
 charge; molecules with one positive charge; mole- 
 cules of ozone (0 3 ) with one positive charge, and 6 
 with one positive charge. Similarly with other gases. 
 In mercury vapour some molecules carry as many as 
 eight positive unit charges. 
 
 Negative charges are found associated only with 
 atoms, never with molecules, and particularly with 
 those of hydrogen and carbon. According to 
 Thomson's conception of the atom, chemical com- 
 bination and valency depend upon the presence of 
 one or more free electrons, so that we get a hint of 
 explanation of the peculiar chemical activity of gases 
 in the ' nascent' state, i.e. at the moment of their 
 formation, before the dissociated atoms have had 
 time to unite into molecules, for it is only then that 
 they can carry a negative charge. 
 
 By these researches a new method of chemical 
 analysis has been opened up, far more sensitive even 
 than the spectroscope. With only one-hundredth of 
 a milligram of a gaseous element present in the tube, 
 and an exposure of less than a millionth of a second, 
 its atomic weight may be determined with an accuracy 
 of one per cent. And not only is the nature of the 
 
 c. 3 
 
34 BEYOND THE ATOM [CH. 
 
 element disclosed, but its physical condition, whether 
 atomic or molecular, charged or neutral. By this 
 means too the greatest difficulty in the accurate 
 fixing of atomic weights, the removal of impurities, 
 is evaded. For impurities make no difference, since 
 each curve corresponds to one and only one type of 
 particle investigated. 
 
 Roentgen or l X' Rays. 
 
 A third type of rays came to light in the late 
 autumn of 1895. Roentgen, repeating Lenard's 
 experiments, had covered the vacuum tube with a 
 black paper case, so that the eyes might be more 
 sensitive, in the complete darkness, for observing the 
 effect of the kathode rays outside the tube. He thus 
 noticed that whenever the tube was working, a 
 fluorescent screen of barium-platinocyanide, which 
 happened to be near it, lit up though no visible light 
 could reach it. On placing the hand between the tube 
 and the screen, to see whether the effect really came 
 from the tube, he saw for the first time the shapes of 
 the bones revealed through the enveloping flesh. 
 
 Though this discovery was in a sense accidental, 
 the fame which it brought was fully deserved. With- 
 in a couple of months Roentgen worked out the con- 
 sequences of his observation so completely that for a 
 year or two after its publication hardly an extra fact 
 

 
 ] THE VACUUM TUBE 35 
 
 was elicited, though all the world precipitated itself 
 on the research, and a sensation that would have been 
 irresistible to a notoriety-hunter was modestly com- 
 municated in due course to the German scientific 
 journals. 
 
 The main properties of the X, or unknown, rays 
 as now established are these. They start from the 
 place where a kathode ray strikes, from the walls of 
 the tube in the first instance, or in a modern ' focus ' 
 tube from the piece of platinum upon which the 
 kathode rays are converged. They issue equally in 
 all directions ; and travel in straight lines, for they 
 cast shadows. Their power of penetration is not 
 selective, as in the case of light, but depends only 
 on the density of the substance penetrated. Hence 
 the opacity of bone and comparative transparency 
 of flesh. In spite of persistent effort no trace of 
 the other familiar properties of light could be 
 detected. The X-rays could be reflected, refracted, 
 polarized, or made to show interference. On the 
 other hand, they are not affected by the most powerful 
 magnetic and electric fields, as the flights of charged 
 particles constituting the kathode and canal rays 
 would be. But they do share with kathode rays 
 the power of exciting fluorescence, affecting a photo- 
 graphic plate, and making the air through which they 
 pass conductive. 
 
 It was this latter property which gradually 
 
 32 
 
36 BEYOND THE ATOM [OH. 
 
 engrossed the attention of scientific investigators, 
 as the novelty of the photographs and shadow 
 pictures of the bones, which had first struck the 
 popular imagination, began to wear off. To account 
 for it, J. J. Thomson and Rutherford brought forward 
 the theory that the rays break up the neutral atoms 
 of the gas into positive and negative carriers, or 
 'ions.' If the space between two plates connected 
 to the positive and negative terminals of a battery 
 is thus i ionized/ the negative ions travel towards 
 the positive plate, the positive ions towards the 
 negative plate, and thus a current passes. But 
 meanwhile the mutual attraction between the positive 
 and negative charges of the ions tends to make 
 them recombine into neutral atoms. The stronger 
 the electric field applied by the battery, the faster 
 the ions will move and the less time will they have 
 for recombination before they reach the plates. The 
 current will therefore increase as the potential 
 difference of the battery is increased, but it will 
 reach a limit when the strength of the battery is 
 sufficient to carry all the ions to the plates before 
 they have a chance of recombining. This limiting or 
 ' saturation ' current affords by far the most accurate 
 and sensitive method of measuring the strength of 
 any ionizing agent such as the X-rays. By its means, 
 and by it alone, has it been possible to detect and 
 discriminate the various radio-active products which 
 
n] THE VACUUM TUBE 37 
 
 exist in quantities so minute as to be far beyond 
 the reach of the chemical balance or even of the 
 spectroscope. Some of them are so shortlived that 
 their existence must be measured in minutes or even 
 seconds ; yet they can be identified with certainty by 
 the ionising effects of their radiations. 
 
 In view of the important conclusions to be based 
 on the theory of ionization it is reassuring to note 
 that many workers, adopting very different methods, 
 have arrived at consistent results concerning the 
 properties of the ions. Their rate of recombination 
 has been measured. Thus if there are a million 
 ions present per cubic centimetre, one half of them 
 will have recombined in less than one second, and 
 99 per cent of them in 90 seconds. The speed of 
 both positive and negative ions, in different gases at 
 different temperatures, has been determined; their 
 coefficient of diffusion ; their approximate size ; and 
 the number produced per second in a cubic centimetre 
 by various agencies. Ionization, it would seem, con- 
 sists in removing an electron from a molecule of the 
 gas. At ordinary pressures the electron at once 
 becomes the nucleus of a cluster of neutral molecules, 
 about thirty in number, the cluster forming a negative 
 ion, through the negative charge of the electron. 
 The rest of the molecule, still practically of molecular 
 size, becomes positive, through loss of the negative 
 charge of the electron. It forms the positive ion, and 
 
38 BEYOND THE ATOM [CH. 
 
 is probably also the nucleus of a cluster of neutral 
 molecules. The conclusion that at ordinary pressures 
 the ions form clusters is deduced from their rates of 
 diffusion. At low pressures they do not form clusters, 
 the negative ion being simply an electron, the positive 
 a particle of atomic size with a positive charge such 
 as is found in the canal rays. 
 
 X-rays lack the characteristic properties of light 
 on the one hand and of streams of charged particles 
 on the other. What then are they? The view 
 generally held was suggested by Stokes and worked 
 out by J. J. Thomson. It is, that they are pulses in 
 the ether originated by the sudden stoppage of an 
 electron in its career. From the negative charge of 
 the electron a Faraday tube of force must reach 
 out through the ether, and it travels with the electron 
 at its high speed, perhaps one-third the speed of light. 
 When the electron is suddenly brought to rest, a jar 
 or pulse travels out along the tube of force, as 
 when a single jerk is given to the end of a stretched 
 rope. Such a pulse would not show reflection, refrac- 
 tion or interference, for these are properties only of 
 a train of waves with a regular period, though it 
 might show a trace of polarization. Yet it is an 
 electromagnetic disturbance of the ether in other 
 respects like light. The more sudden the stoppage 
 of the electron, the thinner will be the pulse, and 
 the more intense the electric and magnetic forces 
 
n] THE VACUUM TUBE 39 
 
 set up in it ; and these must be sufficient to separate 
 an electron from a gas molecule, thus ionizing the gas. 
 The theory explains why the X-rays originate at the 
 point of impact of kathode rays ; and why they are not 
 subject to deflection by an electric or magnetic field. 
 
 But though the ' pulse ' theory of the X-rays at 
 present holds the field, it is by no means yet firmly 
 established. The question of their true nature re- 
 mains one of the outstanding problems awaiting 
 solution. Within the past few years Bragg has 
 brought forward another view, adducing evidence 
 which tends to show that they are corpuscular in 
 nature, like the kathode rays, and not ether-disturb- 
 ances at all. The movements of the ether which 
 we know as light spread out in spheres, like all waves, 
 so that as they recede from the source the intensity 
 falls off according to the well-known law of inverse 
 squares. Now it appears that in an X-ray the energy 
 travels intact. In the picturesque phrase of Stark, 
 who developed the theory from the work of Planck, 
 an X-ray is 'a bundle of energy travelling without 
 alteration of form.' But this is not necessarily incon- 
 sistent with the ' pulse' theory; for if an X-ray is, 
 according to Thomson, 'a kink in the one tube of 
 force which characterises an electron/ the energy 
 would follow the tube, instead of suffering diffusion 
 like ordinary radiations from a source in all directions. 
 
 Other arguments are drawn from the scattering 
 
40 BEYOND THE ATOM [OH. 
 
 of X-rays when they strike upon matter. When 
 X-rays impinge on a thin plate, secondary rays issue 
 from the plate on both sides but in greater numbers 
 on the far side of the plate. These ' secondary' rays 
 turn out to be of the nature of kathode rays, and are 
 therefore corpuscular. Their speed does not depend 
 on the material of the plate, i.e. on the kind of atoms 
 of which it is composed, but does depend on the 
 quality of the X-rays. So that it looks as if they 
 were driven from the plate by the sheer impact of 
 the X-ray, carrying forward, more or less, both its 
 direction and its speed. From a survey of a large 
 amount of work which has been done in recent years 
 on the subject, Bragg concludes that 
 
 'One kathode ray impinging on an atom may 
 produce one X-ray and no more, and in its turn the 
 X-ray through impact on an atom produces one 
 kathode ray and no more, handing on its energy and 
 direction of motion.' 
 
 Such an action is characteristic rather of the 
 impact of material particles than of wave-disturbances. 
 
 But if the X-ray is corpuscular, how shall we 
 account for its extreme penetrating power and its 
 immunity from the action of electric and magnetic 
 fields ? On the first count it cannot be gross matter 
 even of atomic minuteness, it must be electronic in 
 size ; and, on the second, it must be electrically 
 neutral. Thus Bragg concludes ' I have myself found 
 
ii] THE VACUUM TUBE 41 
 
 it convenient to regard the X-ray as a negative 
 electron to which has been added a quantity of 
 positive electricity which neutralizes its charge, but 
 adds little to its mass/ 
 
 And this is his theory of ' neutral pairs.' Against 
 it must be set some researches of Barkla, who has 
 shown that when X-rays of gradually increasing 
 * hardness,' or penetrating power, fall on a given 
 element, suddenly a radiation appears, of the X-ray 
 type, which is emitted equally in all directions and is 
 slightly less penetrating than the rays which excite 
 it. It differs in penetrating power for each element, 
 the penetration usually increasing with the atomic 
 weight of the element, and Barkla has found evidence 
 of several types arising from the same substance, thus 
 forming a characteristic ' spectrum ' for the substance. 
 This is analogous to the fluorescent spectrum emitted 
 by certain substances when light falls on them, and 
 suggests the phenomenon of resonance, as if atoms 
 with one or more natural periods of vibration are 
 only excited when the period of the incident radiation 
 coincides with one of their own natural periods. 
 Moreover Barkla has now found traces of polarization 
 in X-rays reflected from material substances under 
 proper conditions. These facts are consistent with 
 the pulse theory, but hard to account for on any 
 theory that the X-rays are corpuscular in character. 
 
 Thus, though to this day the true nature of the 
 
42 BEYOND THE ATOM [CH. 
 
 X-rays remains an open question, already in 1896 
 the twenty years of study of the high vacuum 
 tube had developed three types of radiation unknown 
 before. The kathode ray had been shown to be a 
 stream of negatively charged particles, minute corn- 
 pared with the smallest atom, but with a speed 
 comparable to that of light itself. The canal rays 
 were also streams of particles, but of atomic size, 
 bearing a positive charge, and having a speed which 
 though still considerable was decidedly less. The 
 X-rays were believed to be not particles at all, but 
 ether disturbances of the nature of light ; differing 
 however from a light ray as a single clash differs 
 from the organised train of waves with a regular 
 period which constitutes a musical note. 
 
 CHAPTER III 
 
 THE NEW RAYS 
 
 To produce X-rays with a vacuum tube electrical 
 energy must be supplied, and then the walls of the 
 tube where the rays originate fluoresce brilliantly. 
 It was natural to suppose tHat the fluorescence was 
 the cause of the rays, and the conservation of energy 
 seemed to demand that they could only be maintained 
 
in] THE NEW RAYS 43 
 
 so long as energy in some form was supplied. Both 
 these clues turned out to be false, and yet they led to 
 a real discovery. 
 
 Many substances were known to fluoresce when 
 excited by the energy of sunlight. Accordingly 
 several investigators Niewenglowski, H. Becquerel, 
 Troost, Arnold exposed fluorescent substances such 
 as calcium sulphide and hexagonal blende to sunlight 
 and found that they gave out rays which could affect 
 a photographic plate through black paper. Le Bon 
 obtained results in this way which he attributed to 
 * dark light.' But all these effects are now known to 
 be due to ultra-violet rays capable, in virtue of their 
 very short wave-lengths, of penetrating substances 
 opaque to ordinary light. 
 
 It was with uranium, well known for its beautiful 
 fluorescence, that the first discovery in radio-activity 
 was made, by Henri Becquerel in February 1896. He 
 found that the double sulphate of uranium and 
 potassium exposed to sunlight could affect a protected 
 photographic plate. The effect could not be due to 
 ultra-violet light, for the rays penetrated thin sheets of 
 metal and glass, which also excluded the possibility 
 of chemical action due to vapours emitted by the salt. 
 Further investigation showed that fluorescence had 
 nothing to do with it. The uranic salts are fluorescent 
 while the uranous salts are not. Yet both proved to 
 be active in proportion to the amount of uranium 
 
44 BEYOND THE ATOM [OH. 
 
 they contained. Nor was the energy of sunlight 
 needed. Salts which had been kept in the dark for 
 years were as active in the dark as when exposed to 
 sunlight. Even crystals which had been deposited in 
 the dark, without ever being exposed to light, showed 
 the same activity. 
 
 The continuous emission of these penetrating 
 rays is thus a specific property of the element 
 uranium itself. It does not depend on the supply of 
 external energy. Nor can the rate of emission be 
 altered by any known physical or chemical process. 
 Lapse of time and extremes of temperature, from 
 that of liquid air upwards, fail to diminish it. 
 Exposure to ultra-violet light or X-rays does not 
 increase it. This was an altogether novel property, 
 and an exhaustive search was at once begun to see if 
 any other known element possessed it. 
 
 It was soon found, by Schmidt and Mme Curie, 
 that thorium was active. Mme Curie then examined 
 the minerals containing uranium and thorium by 
 the electric method, i.e. by their power of ionizing 
 the air between two electrified plates. A striking 
 result followed. Some of the minerals, pitchblendes 
 from the Austrian mines, were about four times as 
 active as uranium itself. Chalcolite, a crystallized 
 phosphate of copper and uranium, was twice as active 
 as uranium, whereas, when the salt was prepared 
 artificially from pure materials, it showed its proper 
 
in] THE NEW RAYS 45 
 
 activity, according to the proportion of uranium 
 contained, i.e. two-fifths of the activity of uranium. 
 The natural mineral was thus five times as active as 
 the artificial, a result that could only be explained 
 by the existence of some impurity far more active 
 than uranium itself. 
 
 The Austrian Government presented M. and Mme 
 Curie with a ton of uranium residues from the State 
 manufactory at Joachimsthal, and there ensued a 
 long and laborious search for the suspected trace 
 of impurity. Endless precipitations and fractional 
 crystallizations were employed, the active agent 
 being unerringly tracked down by testing the 
 activity of the precipitates and residual solutions 
 at every stage, until the patient ingenuity of the 
 investigators was rewarded by the separation of a 
 few milligrams 1 of two active bodies to which Mme 
 Curie gave the names polonium and radium, the first in 
 honour of her native country, the second, by a happy 
 inspiration, on account of its extraordinary power of 
 radiation, more than a million times that of uranium. 
 
 The radium was obtained in the form of a chloride. 
 All there was to show for months of labour on a ton 
 of materials was a minute pinch of white powder, at 
 present worth 15 a milligram or about 1000 a 
 
 1 From a ton of pitchblende may be extracted 200 milligrams of 
 radium chloride, and one five-thousandth as much polonium, i.e. one 
 twenty-fifth of a milligram. 
 
46 BEYOND THE ATOM [CH. 
 
 grain. But its properties amply repaid the toil. Its 
 radiations light up a fluorescent screen brilliantly, 
 blacken a photographic plate, and instantly ionize 
 the air in its neighbourhood. A few milligrams 
 carried in the pocket will soon produce a deep 
 abscess in the side, as M. Becquerel found when he 
 brought some from Paris to London for a lecture at 
 the Royal Institution. 
 
 Since radium is very closely allied to barium in 
 its chemical properties this had been one of the 
 chief difficulties in its separation it was important 
 to make sure that a new element had been discovered, 
 not merely a modified form of barium. This was 
 established when it was found to give a new and 
 characteristic spectrum. Its atomic weight has been 
 determined as 226. It is thus the heaviest known 
 element except uranium. Recently (1911) Mme 
 Curie has at last obtained the pure metal. 
 
 The brilliant researches of the Curies impelled 
 many others to investigate the pitchblende residues. 
 Several active bodies were separated and named 
 actinium, radio-tellurium, radio-active lead. Sub- 
 sequent research has shown that, with the exception 
 of actinium, these substances and also the polonium 
 first discovered by Mme Curie must be referred to 
 the radium family itself, so that the only elements 
 at present known to be radio-active are uranium, 
 radium, thorium, actinium. We shall see that the 
 
in] THE NEW RAYS 47 
 
 first two are members of the same family, while more 
 than a score of new active substances, derived from 
 uranium, thorium and actinium, have been detected. 
 
 What was the nature of these new radiations? 
 Their fluorescent, photographic and ionizing properties 
 could not discriminate them from kathode rays, 
 X-rays, or ultra-violet light, all of which produce 
 similar effects. But each of these properties could 
 be used as a means of measuring their intensity. In 
 this way a further study was made first of their powers 
 of penetration, and then of their susceptibility to the 
 action of magnetic and electric fields. 
 
 It was found at once that the new radiations 
 were not homogeneous. A sheet or two of paper, 
 one or two layers of thin aluminium foil, or even 
 a few centimetres of air, cut off a large part of the 
 effect. There remained a second type of radiation 
 which could only be stopped by a thickness of one- 
 eighth to one quarter of an inch of solid lead. And 
 even then there was a residue which could be detected 
 after passing through six inches of lead or several 
 feet of iron. 
 
 These three types of radiation were named by 
 Rutherford the a, /3, and 7 rays respectively. There 
 are differences in penetrating power among the rays 
 of any one class according to the source. Thus a 
 range of about one ' octave ' in penetrating power is 
 found among a rays, i.e. some a rays are about twice 
 
48 BEYOND THE ATOM [OH. 
 
 as penetrating as the least penetrating a rays. But 
 there is a sharp distinction between the classes. 
 Roughly speaking the penetrating powers of the three 
 types are in the proportion 1 to 100 to 10,000, ft rays 
 being 100 times as penetrating as a rays, and 7 rays 
 again 100 times as penetrating as ft rays. 
 
 When an electric and a magnetic field were 
 applied, it was easily found that the ft rays could be 
 deflected, but the a rays gave much more trouble. 
 Rutherford at last succeeded, and as an indication 
 of the difficulties to be met with in these experi- 
 mental investigations, his arrangement is worth notice. 
 
 To increase the effect, 20 to 25 slits between brass 
 plates from ^ to ^ of an inch apart were employed 
 These were placed beneath the electroscope with a 
 thin layer of radium below them. The electroscope 
 was charged and the rate at which the gold leaf 
 B fell back to the plate D was observed through a 
 mica window by means of a microscope with a 
 micrometer eyepiece. This gave the rate at which 
 the air in the electroscope was ionized by the rays 
 passing up through the slits. It will be seen later 
 that radium gives off a radio-active gas. To prevent 
 this leaking up through the slits an aluminium foil 
 was waxed down above them and a current of hydro- 
 gen was kept flowing down through the porous foil. 
 The most powerful magnetic field available was then 
 applied so as to deflect the rays towards the walls 
 
HI] 
 
 THE NEW RAYS 
 
 49 
 
 of the slits. The a-rays could be cut off by covering 
 the radium with a sheet of mica, which allowed the 
 & and 7 rays to pass. The ionization could thus be 
 observed (1) with or without the a rays, (2) with or 
 without the magnetic field. The results showed that 
 
 Earth 
 
 -Inflow of Hydrogen 
 
 Eos. 
 
 utflow of Hydrogen 
 Fig. 2. 
 
 the a rays were deflected. By covering half the top of 
 each slit by a brass ledge projecting from one side it 
 was possible to detect the direction of the deviation. 
 Later on the rays were received on a photographic 
 plate and the amount of deflection measured. The 
 way was thus open to apply to the a and ft rays the 
 
 c. 
 
50 BEYOND THE ATOM [CH. 
 
 same methods by which the velocity and ratio of 
 charge to mass had been determined for the kathode 
 and canal rays. 
 
 The general results obtained were as follows. 
 
 The a rays bear a positive charge and the particles 
 of which they consist have a speed about one-tenth 
 the velocity of light. The ratio of charge to mass is 
 about half that of the hydrogen atom in electrolysis. 
 Though deviable by magnetic and electric fields, the 
 amount of deviation is minute compared with that of 
 kathode rays or 13 rays. They thus correspond to the 
 particles of the canal rays in the vacuum tube, being 
 of atomic size about twice the mass of the hydrogen 
 atom. 
 
 The /3 rays consist of particles with a negative 
 
 p 
 
 charge. The ratio - is 10 7 , the same as in the 
 
 HYl 
 
 kathode rays. The velocity ranges from about two- 
 fifths up to nine-tenths the velocity of light itself. 
 The particles of the ft rays are thus identical with 
 those of the kathode ray, being about one-thousandth 
 of the mass of the hydrogen atom. They are in fact 
 electrons, but with an enormously greater speed than 
 those projected in the vacuum tube. 
 
 The 7 rays cannot be deflected by magnetic or 
 electric fields. They resemble in all respects the 
 X-rays, but are far more penetrating than X-rays 
 from the ' hardest ' vacuum tube. They are probably 
 
Ill] 
 
 THE NEW RAYS 
 
 51 
 
 of the same nature, i.e. not particles at all, but thin 
 pulses in the ether. 
 
 It thus appears that certain of the heaviest ele- 
 ments, uranium, radium, thorium, 
 are continuously projecting, un- 
 aided, without any supply of 
 external energy, three types of 
 radiation strictly analogous to the 
 three types which may be pro- 
 duced from a vacuum tube driven 
 by a powerful induction coil. 
 They may justly be called ' radio- 
 active' elements. 
 
 The fact that the a particles 
 carry a positive and the /3 par- 
 ticles a negative charge has been 
 deduced above from the direction 
 in which they are deflected by a 
 magnetic field. This has been 
 proved directly, though the ex- 
 perimental difficulties were great. 
 This chapter may fitly be closed 
 with a reference to Strutt's ' ra- 
 dium clock,' at once a demon- 
 stration and an ingenious appli- 
 cation of the fact. 
 
 An active preparation of thirty 
 milligrams of radium bromide are Fig 3 
 
52 BEYOND THE ATOM [CH. 
 
 sealed into a thin glass tube A, which is supported 
 on an insulating quartz rod B within an exhausted 
 outer tube D. The walls of A are made conducting 
 by a thin coating of phosphoric acid. From the lower 
 end of A two gold leaves CO are suspended, in metallic 
 connection with the radium in A. 
 
 The walls of A absorb all the a particles discharged 
 from the radium, but allow the /3 particles to escape 
 to the outer tube. The radium and the leaves 
 attached to the tube thus acquire a positive charge 
 and the leaves steadily diverge. When they come in 
 contact with strips of tin foil EE, connected with the 
 earth, the leaves are discharged and collapse, only to 
 begin at once the same process over again. This 
 operation is repeated about once a minute, and from 
 what is now known it would still be in progress, 
 though only half as quickly, after the lapse of two 
 thousand years. This is the nearest approach to 
 perpetual motion hitherto attained. 
 
 CHAPTER IV 
 
 THE NEW SUBSTANCES 
 
 URANIUM and thorium, in view of their strange 
 property of radio-activity, were naturally subjected 
 to much investigation. In the early days of this 
 
iv] THE NEW SUBSTANCES 53 
 
 research some puzzling and indeed at the time 
 inexplicable facts came to light. 
 
 For instance, all the evidence pointed to the 
 conclusion that radio-activity was a specific property 
 of the atom of uranium or thorium. The activity of 
 any salt was proportional simply to the amount of 
 element it contained, and depended in no way on its 
 state of chemical combination. No known chemical 
 or physical agency was able to affect the rate at 
 which the process went on. In 1900, however, Crookes 
 precipitated a uranium solution with ammonium 
 carbonate, dissolved the precipitate in excess of the 
 carbonate, and finding that a light precipitate re- 
 mained, filtered it out. This residue, though chemically 
 free from uranium, proved to be several hundred times 
 as active as uranium when tested photographically, 
 while the uranium solution had lost the power of 
 acting on a photographic plate. 
 
 Rutherford and Soddy then attacked thorium, 
 and by precipitating it with ammonia obtained a 
 residue several thousand times as active as thorium, 
 though chemically free from it. The precipitated 
 thorium was found to have lost more than half its 
 activity. 
 
 The active substances thus separated were called, 
 from their 'unknown' nature, Uranium X and 
 Thorium X. That they were new substances with 
 distinct chemical properties seemed clear from the 
 
54 BEYOND THE ATOM [OH. iv 
 
 way in which they had been obtained as the result 
 of specific chemical processes. For the ammonium 
 carbonate, which separated uranium X, fails with 
 thorium ; and ammonia, which was at the time the only 
 agent capable of completely precipitating thorium X, 
 fails to separate uranium X \ Thus new substances had 
 been separated from uranium and thorium possessing 
 the property which seemed to belong to the atoms 
 of those elements, and leaving the latter partially or 
 wholly inert. 
 
 A still more remarkable fact was to follow. 
 Becquerel set aside an active solution of Ur X and 
 the uranium from which it had been obtained. On 
 examining them at the end of a year he found that 
 the uranium solution had regained its activity, while 
 that of the Ur X had disappeared. The same was 
 true for thorium solutions. Rutherford and Soddy 
 found that at the end of a month the thorium X was 
 no longer active, while the thorium had regained its 
 activity. Obviously the next step was to trace in 
 detail the laws of decay and recovery with lapse of 
 time. Rutherford and Soddy did this in a long series 
 of measurements for thorium and then for uranium. 
 
 1 Several other agents are now known to be capable of precipita- 
 ting thorium X, but the success of Crookes' experiment with 
 ammonium chloride has been found to depend on the presence of a 
 trace of impurity ! So that uranium and Ur X are the Dr Jekyll and 
 Mr Hyde of the subject ! One way is to add a trace of iron ferric 
 hydroxide to the pure salt. 
 
56 BEYOND THE ATOM [OH. iv 
 
 Fig. 4 shows the results for uranium and Ur X plotted 
 in curves. 
 
 Consider first the curve of decay of Ur X. It 
 is a curve well known to mathematicians as an ex- 
 ponential or logarithmic curve. It indicates a law 
 frequently met with, viz. that the quantity changing 
 at any instant is a fixed fraction of the total amount 
 existing at that moment. When, for instance, a 
 monomolecular chemical change is taking place, i.e. 
 a change in which only one kind of matter is involved, 
 it is according to this law that it proceeds. Of all 
 the unchanged molecules existing at the beginning 
 of any second a fixed proportion will effect the 
 change during that second. The change therefore 
 begins fast, proceeds more and more slowly as time 
 elapses, and theoretically should never reach com- 
 pletion. The swings of a pendulum left to itself die 
 away according to this law. The energy taken out of 
 the pendulum by frictional resistances during each 
 swing must be proportional to the length of the swing 
 during which they can act. Hence each swing is less 
 than the last by a fixed fraction of the last. This, 
 too, is the law of decay of an electric current in an 
 inductive circuit when the source of power is suddenly 
 cut off. The processes of radio-activity cannot be 
 understood unless this principle is grasped. Therefore 
 a more familiar illustration may be given. If the 
 decay curve is taken in the reverse direction, beginning 
 
58 BEYOND THE ATOM [OH. 
 
 with the small value in the right-hand bottom corner, 
 it represents the growth of a sum of money invested 
 at compound interest ; for the rate of increase at any 
 moment is a fixed percentage of the amount existing 
 at that moment. 
 
 All logarithmic curves are of the same type and 
 are expressed by a mathematical equation of the 
 same form. They only differ in the rate at which 
 decay takes place. This is most conveniently measured 
 by the time taken for the original amount to fall to 
 half its value. The moment this is known, the whole 
 curve can be drawn. In the case of uranium X it is 
 about 22 days. For thorium X (fig. 5) it is about 
 4 days. Otherwise, as the figures show, the curves 
 are precisely similar. 
 
 Turning now to the recovery curve, we see that 
 it is nothing but the decay curve upside down. It 
 rises rapidly at first, then more and more slowly, 
 approaching the normal value of 100, but never 
 reaching it, precisely as the decay curve approaches 
 zero. The growth in any second is & fixed fraction 
 of what still remains to be accomplished. And here 
 a very important point must be noted. The sum of 
 the values for the two curves at any one instant is, 
 if the curves are accurately drawn, exactly 100, and 
 equal to the original amount. This is only approxi- 
 mately true in the figures given, which represent 
 actual experiments liable to slight errors. The two 
 
iv] THE NEW SUBSTANCES 59 
 
 curves are in fact complementary, the total activity 
 existing in the uranium X and the uranium being 
 precisely the same throughout the process. As the 
 Ur X loses its activity, step by step the uranium 
 regains what the other loses, as if by direct transfer 
 between them. This is found experimentally to be 
 the fact, whether the two solutions are kept near 
 each other, or separated to great distances, or enclosed 
 in lead which would prevent any possible direct 
 radiation. Nor can any known agency affect the 
 rate of the process. Extremes of temperature, light 
 and darkness, vacuum and high pressure, violent 
 chemical reagents and electric forces have all been 
 tried, but the decay and recovery go irresistibly on. 
 Strange as these facts appeared at the time, they 
 admit of a simple explanation. But before it was hit 
 upon, two similar phenomena had been discovered. 
 
 The Emanations. 
 
 Worlrers with thorium had been much troubled 
 by unaccountable irregularities in their measurements 
 of radiation. When an open testing vessel was used, 
 a slight current of air, or the mere opening of a door 
 at the other end of the room, would cause great 
 differences in the readings of the electrometer. On 
 investigation Rutherford concluded that a minute 
 trace of some active substance was given off by 
 thorium, which could be blown about from place 
 
60 BEYOND THE ATOM [CH. 
 
 to place. Whether this was a vapour, or perhaps a 
 fine dust, such as that to which the odour of musk 
 and some metals is due, or, may be, even a new gas, 
 was doubtful. In order not to prejudge the question 
 Rutherford happily called it the 'emanation.' It was 
 soon found to possess many of the properties of an 
 active gas. It diflused through paper and thin metal 
 foil, but not through mica or glass. It could be 
 carried by a current of air through tightly packed 
 cotton wool, and bubbled through solutions without 
 losing its activity. When a quantity of the emana- 
 tion was left to itself, however, its activity decayed 
 by an exponential law, just as the action of Ur X 
 and Th X do, only much more rapidly. The activity 
 of the thorium emanation decays to half value in 
 about one minute. On the other hand, the activity 
 of the emanation due to a thorium compound placed 
 in a closed vessel rises to a maximum, reaching half 
 value in about one minute. The two curves are 
 complementary, exactly as in the case of uranium 
 and uranium X. 
 
 When radium was examined it was also found to 
 give out an emanation which behaved in the same 
 way except that it had a longer life. The activity 
 of the radium emanation, left to itself, decays to half 
 value in a little less than four days, about 5000 times 
 the time taken by the thorium emanation. 
 
 Evidence accumulated that these emanations were 
 
iv] THE NEW SUBSTANCES 61 
 
 new gases of high atomic weight akin to the argon 
 family. They behaved like ordinary gases in diffusion 
 experiments, and their coefficients of diffusion in- 
 dicated high atomic weight. They passed unscathed 
 through a platinum tube heated electrically to the 
 highest possible temperature and filled with platinum 
 black; through red hot lead chromate, magnesium 
 powder, zinc-dust, reagents which would have affected 
 any known substances except the heavy gases of the 
 argon family. 
 
 But the real proof of their gaseous nature was 
 only effected when Rutherford and Soddy succeeded 
 in condensing them by means of low temperatures. 
 A liquid-air machine was specially obtained for this 
 experiment, and the writer well remembers how with 
 the first litre of liquid air produced the condensation 
 was successfully brought about. The emanation was 
 conveyed to the testing vessel by a slow stream of 
 hydrogen passing on its way through a spiral which 
 could be immersed in liquid air. So long as the 
 spiral was immersed, no trace of activity was seen 
 at the electrometer. But if the liquid air was re- 
 moved and the spiral allowed to warm up, at a 
 temperature about 154 C. below zero the needle of 
 the electrometer suddenly moved through several 
 hundred divisions, showing that the accumulated 
 emanation which had been condensed during im- 
 mersion had been set free. 
 
62 BEYOND THE ATOM [CH. 
 
 The emanating power of the various compounds 
 of thorium and radium, i.e. the rate at which they 
 give off emanation, depends much on chemical and 
 physical circumstances. Thus thorium nitrate in 
 solution gives off 600 to 800 times as much as in 
 the solid state. But Rutherford and Soddy have 
 shown that these differences are due to the occlusion 
 of the gas in the solid. The rate of production is 
 independent of circumstances. They also proved that 
 the origin of the thorium emanation is not thorium, 
 but thorium X. When Th X is separated from a 
 thorium solution, the Th X has marked emanating 
 power, while the precipitated thorium hydroxide 
 shows scarcely a trace. But now the Th X begins 
 to lose its emanating power, while the hydroxide 
 regains it; and if the curves are plotted, they prove 
 to be identical with the curves of decay and recovery 
 of activity for Th X and the thorium hydroxide. 
 The active radiation of Th X thus accompanies the 
 emission of the emanation from it. 
 
 Throughout the elaborate researches summarized 
 in this chapter the only guide has been the activity 
 of the substances concerned, measured either photo- 
 graphically or by the electric method. The quantities 
 obtained are too small to be detected otherwise. 
 Ur X and Th X are mainly impurities associated 
 with uranium and thorium, containing an infinitesimal 
 trace of the active constituent. So, too, with the 
 
iv] THE NEW SUBSTANCES 63 
 
 emanations. The amount of thorium emanation 
 is too small and it decays too rapidly ever to be 
 detected by its volume. The rate of production 
 of emanation by radium is about a million times 
 greater, and it has a longer life. Nevertheless the 
 emanation from one gram of radium is less than 
 one cubic millimetre. Yet Ramsay and Soddy by 
 extraordinarily skilful experiment, with only 60 milli- 
 grams of radium at their disposal, not only detected 
 and measured the volume, but proved that the minute 
 bubble of gas obeyed Boyle's law, and showed that 
 its volume, when left to itself, decreased according to 
 the same law by which the activity of the emanation 
 was known to decay. ft is now known that even of 
 this bubble only 5 per cent, was pure emanation. 
 The first accurate measurement was made by 
 Rutherford, who purified the emanation and obtained 
 only one-twentieth of the amount found by Ramsay 
 and Soddy. 
 
 The Active Deposits. 
 
 It was discovered (by M. and Mme Curie for 
 radium and by Rutherford for thorium) that solids 
 which have come in contact with the emanations 
 become themselves radio-active. They behave as if 
 active matter had been deposited upon them from 
 the emanations. 
 
64 
 
 BEYOND THE ATOM 
 
 [CH. 
 
 This effect is due not to the radiations, but to 
 actual contact with the emanations, as was proved 
 by Rutherford, from an experiment due to P. Curie. 
 
 A solution of a radium salt is placed inside a closed 
 vessel within which various plates A, B, C, D, E (fig. 6) 
 are disposed, some protected from direct radiation by 
 heavy lead plates, as D, others like E, directly exposed. 
 
 Fig. 6. 
 
 All become radio-active, the intensity being inde- 
 pendent of the position or material of the plates, and 
 proportional simply to their area. If the solution is 
 covered with a few sheets of paper, which cut off the 
 a rays but allow the emanation to diffuse through, 
 the plates become active. If a sheet of mica is waxed 
 
iv] THE NEW SUBSTANCES 65 
 
 down over the solution, thus stopping the emanation, 
 no activity is produced. Uranium and polonium, 
 which do not produce an emanation, do not render 
 the plates active. If one of the plates is connected 
 to the negative terminal of a battery or electric 
 machine, the whole of the activity will be concentrated 
 on that plate. 
 
 In this way a short thin platinum wire, negatively 
 charged and exposed to the thorium emanation, may 
 be made more than 10,000 times as active as the 
 thorium from which it was derived. 
 
 The active deposit may be rubbed off the wire 
 or plate by sand paper, but then it is found on the 
 paper. It can be driven off by electrically heating 
 the wire, but is then found on the interior of the 
 vessel in which the wire is heated, as if it had been 
 sublimed. It has definite chemical properties; for 
 it will not dissolve in water or nitric acid, but is 
 soluble in sulphuric and hydrochloric acids. When 
 the acid solutions are evaporated, the activity remains 
 on the dish. 
 
 It is thus clear that the emanations give rise to 
 a solid active deposit of definite chemical character. 
 But the amount must be infinitesimal ; for not the 
 slightest difference in weight can be detected, nor 
 can anything be seen on the active wire under the 
 most powerful microscope. 
 
 When a wire which has been exposed for a long 
 
BEYOND THE ATOM 
 
 [CH. 
 
 time to thorium emanation is left to itself, its activity 
 decays in the now familiar way by an exponential 
 law, the characteristic time of falling to half value 
 being about 11 hours. As we are now prepared to 
 
 100 
 
 40 60 
 
 Time in Hours. 
 Fig. 7. 
 
 80 
 
 100 
 
 expect, the curve showing the rise of activity in the 
 wire during exposure to the emanation is the com- 
 plementary curve. The two are shown in fig. 7. 
 We have now traced the discovery and the laws of 
 
v] DISINTEGRATION 67 
 
 production and decay of the following new substances, 
 new because absolutely unsuspected by chemists 
 before their existence was revealed by their radio- 
 active properties : 
 
 Uranium X. 
 
 Thorium X The Emanation Active Deposit. 
 
 Radium Emanation Active Deposit. 
 Uranium X produces no emanation. On the other 
 hand radium gives rise directly to the radium 
 emanation, without the intervention of anything 
 corresponding to uranium X and thorium X. 
 Thorium alone gives the complete series, 
 
 Thorium Th X The Emanation Active Deposit, 
 
 each member of which seems to be the parent of the 
 next. 
 
 This was the stage at the end of 1904. We shall 
 see that a great development has taken place since. 
 
 CHAPTER V 
 
 DISINTEGRATION 
 
 THE complicated facts detailed in the last two 
 chapters, together with many others too numerous to 
 be included within the limits of this essay, have 
 received a beautifully simple explanation, now to be 
 
 52 
 
68 BEYOND THE ATOM [CH. 
 
 described, which links them all together, each in its 
 natural place, as parts of a single theory. Yet this 
 theory ran so counter to received ideas that it was 
 not without conflict it attained its present universal 
 acceptance. 
 
 Two main facts had to be explained. First, the 
 production from apparently inert matter of the new 
 radiations, representing a very large output of energy 
 compared with the weight of the substances from 
 which it was emitted. And next, the simultaneous 
 production, proceeding step by step with the radia- 
 tions, of a whole series of new substances with dis- 
 tinctive chemical properties, themselves radio-active, 
 with an activity, however, which decayed according 
 to characteristic laws. 
 
 Looking back from our present point of vantage, 
 it is hard to realize the confusion introduced by the 
 new facts among the ideas current ten years ago, and 
 the apparent shock to the principles then accepted as 
 established. The century-long belief in an unchange- 
 able indestructible atom forbade anything more than 
 a chemical reaction between elements ; confidence in 
 the conservation of energy compelled a search for the 
 source of the energy expelled, which enormously ex- 
 ceeded, weight for weight, the energy disengaged in 
 the most powerful chemical reaction known, that 
 between oxygen and hydrogen. Small wonder that 
 the investigators cast about rather wildly for an idea 
 
v] DISINTEGRATION 69 
 
 that would fit the facts. The Curies suggested that 
 the emanation was not material, but consisted of 
 'centres of condensation of energy attached to the 
 gas molecules ' ; that the radio-active particles had 
 the power of abstracting energy from the surrounding 
 air, contrary to Carnot's principle ; or again, that they 
 obtained it by absorbing Roentgen rays passing 
 through the atmosphere. Kelvin, as late as 1903, 
 suggested that radium could absorb some unknown 
 radiations. Crookes held that the energy was ab- 
 stracted from the gas, the radio-active particles 
 receiving the impact of a gas molecule and sending it 
 out with a lowered velocity. 
 
 The theory which has cleared away all difficulties 
 is known as the Disintegration Theory. It was 
 brought forward by Rutherford and Soddy in 1902, 
 for thorium, and in 1903 for radium and radio-active 
 change in general. It consists of two parts : first, an 
 interpretation of fact, as revealed by the curves of 
 production and decay of activity of the new sub- 
 stances : and next a suggestion, as bold as it is simple, 
 of the physical mechanism involved in the changes. 
 
 Interpretation of the Curves. 
 
 Consider the two complementary curves which 
 express the growth and decay of activity in any of 
 the cases described in the last chapter, for instance, 
 
70 BEYOND THE ATOM [OH. 
 
 those for uranium and Ur X (fig. 4). These curves 
 indicate 
 
 (1) that uranium produces an active substance 
 Ur X at a constant rate, unalterable by any known 
 process ; 
 
 (2) that Ur X loses its activity according to an 
 exponential law. 
 
 The activity of uranium which has been left to 
 itself for a long time is thus due to an equilibrium 
 between the constant production of fresh Ur X, and 
 the gradual loss of activity of the Ur X already 
 existing. 
 
 When the Ur X is separated from the uranium, it at 
 once begins to lose its activity according to the decay 
 curve. Meanwhile the uranium goes on producing 
 fresh Ur X at the same rate, and thus begins to 
 regain its activity. The fresh Ur X loses its activity, 
 like that already separated, but the constant pro- 
 duction of new Ur X increases the activity of the 
 uranium solution till so much has been produced that 
 the fractional loss of activity of the new Ur X is 
 again just equal to the activity of the new Ur X pro- 
 duced per second. Equilibrium is thus again estab- 
 lished and the uranium is found to have regained its 
 normal activity. 
 
 One feature of the complementary curves is at 
 once explained. Since the rate of loss of activity of 
 the Ur X is unaltered by separation, and the new 
 
v] DISINTEGRATION 71 
 
 Ur X falls off at the same rate as if separated, the 
 result would be the same if the two solutions were 
 kept together, and the total activity is unchanged 
 throughout. Hence, as is found from the curves to 
 be the case, the sum of the activities for the two 
 curves is always the same, and equal to the normal 
 activity. 
 
 The importance of this principle justifies a familiar 
 illustration. Let us suppose a man falls heir to an 
 estate from which 100 a month is paid into his ac- 
 count, and resolves to spend each month one tenth of 
 all he has in the bank. His account for the first year 
 will run as follows : 
 
 Months In Bank Spends Balance 
 
 1 100 10 90 
 
 2 190 19 171 
 
 3 271 27 234 
 
 4 334 33 301 
 
 5 401 40 361 
 
 6 461 46 415 
 
 7 515 51 464 
 
 8 564 56 508 
 
 9 608 61 547 
 
 10 647 65 582 
 
 11 682 68 614 
 
 12 714 71 643 
 
 The reader is advised to plot these numbers on 
 squared paper, and continue them further. He will 
 obtain the curve of growth of fig. 4. It is clear 
 
72 BEYOND THE ATOM [CH. 
 
 that the amount in bank increases fast at first, then 
 more slowly, and tends at last to reach 1000, though 
 it never actually attains that sum. If it were attained 
 there would be equilibrium. For then 100 would 
 be spent each month, and this would be replaced 
 exactly by the 100 regularly paid in. 
 
 It will be noted that the increase in the amount 
 in bank at the beginning of each month is one-tenth 
 of what remains to be gained in order to reach tlie 
 1000. Thus in the first month 900 is required and 
 the increase is 90. This was pointed out as a property 
 of the curve of growth. 
 
 If, now, after a year or two, when the normal 
 1000 has practically been attained, the trustees 
 are temporarily unable to continue their payments, 
 and the man starting with his 1000 goes on living 
 by the same rule, he will spend 100 the first month, 
 90 the second, 81 the third and so on. His capital 
 will fall off according to tJie decay carve and approach 
 zero. 
 
 But should the trustees, foreseeing that their 
 difficulty is only a temporary one, authorize him to 
 count on the ultimate reception of the 100 a month, 
 his expenditure, i.e. the total activity, would remain 
 at 100, being drawn less and less from the 1000 in 
 hand, and more and more from the accumulating 
 hundreds not immediately available. By the time 
 he would have exhausted his 1000, the normal 
 
v] DISINTEGRATION 73 
 
 1000 would just be re-established in his bank 
 account. 
 
 The Physical Mechanism. 
 
 Up to this point there is nothing hypothetical. 
 We have merely interpreted the meaning of the curves 
 and learned from them the facts as to the production 
 of the new substances. It remains to guess the 
 physical process which underlies these changes ; and 
 it must be one which will at the same time account 
 for the emission of the radiations and the enormous 
 output of energy which accompanies them. 
 
 The daring suggestion put forward by Rutherford 
 and Soddy was that in every second a certain per- 
 centage of the existing atoms of uranium break up or 
 disintegrate, each atom throwing out an a particle 
 with great velocity. The mass of an a particle, from 
 
 the value of its ratio - , appeared to be twice that of 
 m' 
 
 a hydrogen atom ; but there was evidence indicating, 
 what has since been proved, that the a particle is 
 really an atom of helium, with four times the mass of 
 a hydrogen atom, only carrying a double charge, 
 
 p 
 
 which would give the same value for . The atomic 
 
 m 
 
 weight of uranium is 238. On the expulsion of the a 
 particle 4, the residue would be an atom 234 l compared 
 
 It has been found since that uranium emits two a particles, so 
 that the atom of Ur X is 230. See Table p. 91. 
 
74 BEYOND THE ATOM [OH. 
 
 with hydrogen, and this residue is an atom of uranium 
 X. The new atom is itself unstable, and every second 
 a certain percentage of those existing will in turn 
 break* up, ejecting in this case a P and a j ray. Since 
 the y ray is not material, and the y3 ray is only an 
 electron, the residue is still an atom of Ur X, but 
 lacking an electron. For uranium the next stage 
 could not then be traced. 
 
 Why then, the reader may ask, does uranium lose 
 its activity on the separation of Ur X, though it goes 
 on producing Ur X at the same rate, ejecting an a 
 particle for every atom formed? This is because 
 Crookes used the photographic method in testing the 
 activity, and the a rays of uranium have very little 
 photographic effect, which is due almost entirely to 
 the /3 rays. On the other hand, the a rays do most 
 of the ionizing, and had he used the electrical method, 
 he would not have found the activity of the uranium 
 much diminished. 
 
 . Two other points should be noted : in any series 
 of radio-active changes the /3 rays do not in general 
 appear until the last that can be traced, or at all 
 events the last preceding either a rayless> or on other 
 grounds presumably less violent change. For example, 
 the last products of the Thorium and Actinium series 
 break up with such violent explosions, and Radium (7, 
 the last of the rapid changes in the radium series, ends 
 with one. The ft particles are shot out with astounding 
 
v] DISINTEGRATION 75 
 
 violence, -having a speed often closely approaching 
 that of light itself. Those from radium C attain 
 "998 of the speed of light. The cause of the disruption 
 of an atom is not known, but it looks as if the fi 
 particle is the disturber, and once it is got rid of 
 comparative peace is restored, though several a par- 
 ticles may be thrown out before this can be brought 
 about. 
 
 The 7 ray only and always accompanies the fi 
 ray. This is what should be expected, if 7 rays are 
 pulses in the ether generated by the stoppage, and 
 therefore also by the sudden starting of a (3 particle. 
 
 The changes that take place in thorium and 
 radium are far more complicated than in the case 
 of uranium; but a vast amount of patient experi- 
 menting has succeeded in elucidating them. We will 
 only note here the first few steps disclosed. 
 
 Thorium emits an a particle at each of the changes 
 from thorium to thorium X, from thorium X to the 
 emanation, and from the emanation to the active 
 deposit. The active deposit gives out all three types 
 of radiation, a, ft and 7 rays. Radium gives out a 
 particles on changing to the emanation, and again 
 when the emanation passes into the active deposit. 
 The deposit gives all three types. 
 
 The theory thus explains why the transformations 
 are accompanied, step by step, by the radiations. It 
 also suggests the source of the energy developed. 
 
76 BEYOND THE ATOM [CH. 
 
 It must be withdrawn from the internal energy of the 
 atom. The atom is to be conceived as a complex 
 planetary system in which negative electrons and 
 positive a particles, or some other form of positive 
 electricity, revolve about each other with enormous 
 speed. For some as yet unknown reason some of the 
 atoms of the radio-active elements from time to time 
 become unstable, ejecting an a particle with its orbital 
 velocity, just as a stone leaves a sling, or the parts of 
 a fly wheel are projected, if the speed becomes too 
 great for the strength of the spokes. 
 
 With a natural reluctance to admit the possibility 
 of change in an atom, it has been suggested that 
 radium is but a chemical combination of helium and 
 the emanation ; and so for the other cases of radio- 
 active change. But the reasons which compel us to 
 regard radio-active changes as atomic, not molecular, 
 are too strong. The spectra of radium, the emanation, 
 helium are absolutely distinct and characteristic, as 
 of new elements, instead of being related as they 
 should be for a compound and its components. The 
 energy released is more than a million times greater 
 than in the most intense chemical reaction known 
 between molecules. Chemical reactions are greatly 
 influenced by heat and other physical agents such 
 as electric force. But radio-active changes cannot 
 be accelerated or retarded by any known physical 
 agency. This would be strange if they were reactions 
 
v] DISINTEGRATION 77 
 
 between molecules as in chemistry. But it is just 
 what would be expected if they are changes in the 
 atom, for the very name element indicates that the 
 forces employed in ordinary chemistry have failed 
 to produce any further effect. Finally, no known 
 chemical reaction gives rise to the radiations that 
 accompany radio-active change. 
 
 It is not surprising that this theory met with 
 opposition. The wonder is that so profound a 
 revolution in ideas should have been effected within 
 a single decade. The unchangeable atom is seen in 
 the act of dissolution. The alchemist's dream of 
 transmutation of the elements is in spontaneous 
 operation from day to day. Not, indeed, between 
 the nobler and baser metals, nor is the process 
 subject to the experimenter's control. ' II n'y a pas ' 
 says Mme Curie in her recent monumental work on 
 Radio-activity (1911) ' encore actuellement de raisons 
 suffisantes pour admettre que la formation de certains 
 elements puisse 6tre provoque*e t\ volonte en presence 
 de corps radio-actifs. La production d'he'liurn reste 
 acquise ; mais elle est relie'e k une propriety essentielle 
 des elements radio-actifs et n'est pas influencee par 
 1'intervention de I'experimentateur.' 
 
 The reason why the new views made way so 
 readily is that the ground was already prepared. It 
 often happens that a step in scientific discovery 
 must wait a century or so after its first prevision 
 
78 [BEYOND THE ATOM [OH. 
 
 before it can be taken, because progress in some 
 other part of the field lags behind the general 
 advance. But in the case of radio-activity all was 
 ready. The three types of radiation emitted by the 
 radio-active elements had been elucidated by the 
 study of the vacuum tube. Thomson had demolished 
 the solid unchangeable atom and discovered the 
 electron. The X rays had brought about the ioniza- 
 tion theory of the conduction of electricity through 
 gases, on which the best method of making measure- 
 ments in radio-activity is based. It was a singularly 
 fortunate coincidence that the leading ideas required 
 for both theory and measurement should be thus to 
 hand. As early as 1900 Mme Curie, in her thesis 
 for the doctorate, had spoken of ' transformations not 
 ordinary but atomic/ of the violent interior movement 
 in the atoms, which she described as unstable, l en train 
 de se disloquer,' and radiating particles less than an 
 atom, while the sub-atoms were in movement. But it 
 was in 1902 that Rutherford and Soddy first gave the 
 theory a definite form and showed in detail its aptitude 
 for explaining the facts known in the case of thorium. 
 
vi] A FAMILY TREE 79 
 
 CHAPTER VI 
 
 A FAMILY TREE 
 
 RADIO-ACTIVE change is by no means complete 
 when the active deposit is reached. To disentangle 
 the complicated subsequent changes has been the 
 object of a vast amount of ingenuity and patient 
 labour. It is marvellous that so much has been 
 accomplished when the total quantity of matter 
 available for work was far too small to be detected 
 by the balance, the microscope, or the spectroscope. 
 In default of the usual means employed by the 
 chemist, investigators had at their disposal the 
 three methods of studying the radiations, by their 
 photographic, fluorescent and, above all, their ionizing 
 effects. When radiation fails, the further transfor- 
 mations, if they exist, pass out of our ken. As a 
 type of this work let us consider what has been 
 discovered of the radium family. 
 
 Radium, as we have seen, does not produce 
 Radium X, but with the emission of an a particle 
 yields the emanation direct. With a solid salt of 
 radium the emanation is mostly occluded in the salt. 
 But it is disengaged on solution. Accordingly a 
 radium solution is placed in a flask, and a platinum 
 
80 BEYOND THE ATOM [CH. 
 
 wire, maintained negative, is inserted through a rubber 
 cork. The emanation decays to half value in about 
 four days, and forms the active deposit on the wire, 
 again with the emission of an a particle. 
 
 The wire can be removed at any stage, and its 
 activity studied. It is found that the greater part of 
 this activity disappears within 24 hours ; but a very 
 small part, less than one hundred thousandth of the 
 whole, remains and goes through a much slower series 
 of changes. 
 
 Take first the activity of rapid change. The 
 curves of decay obtained are quite different accord- 
 ing as it is (1) tested by the a, or the /3 and 7 rays 
 (the latter give identical curves, since the 7 ray always 
 accompanies the y8 ray) ; and (2), exposed for a long 
 or a short time to the emanation before it is examined. 
 
 Figure 7 shows the curves for the a rays. Both 
 curves exhibit three well-marked stages. The first 
 and the last stages follow an exponential law, the 
 times of decay to half value being about 3 minutes 
 and 28 minutes respectively. The middle stage is 
 not exponential, but indicates a conflict between a 
 substance decaying exponentially, and fresh produc- 
 tion of that substance, not however at a constant rate. 
 When mathematical equations are applied to these 
 curves, it is found that they can be completely 
 explained by assuming three successive changes: 
 
 (1) The emanation produces a substance (radium 
 
VI] 
 
 A FAMILY TREE 
 
 81 
 
 A) which decays rapidly and is half transformed in 
 3 minutes, emitting a rays. 
 
 (2) The matter derived from radium A, called 
 radium B, is half transformed in 21 minutes, but gives 
 out no ionizing (a) rays during the process. 
 
 Decay of Excited Activity of Radium 
 measured by a rays. 
 
 10 
 
 20 
 
 30 40 
 
 Time in Minutes. 
 
 50 
 
 CO 
 
 Fig. 8. 
 
 (3) The matter derived from radium B, called 
 radium C, is half transformed in 28 minutes, and gives 
 out a rays. 
 
 The sudden drop at the beginning is thus due to 
 c. 6 
 
7 
 
 
 88? 
 
 otjmpjf /o fiji 
 
 g3 oJ 
 
 5 . 
 
 ^ &t4 
 
 fi 
 
CH. Vl] 
 
 A FAMILY TREE 
 
 83 
 
 the rapid decay of radium A. In the middle stage 
 the decay of radium C already produced is masked 
 for a time by fresh production of radium C. When 
 this has practically ceased, the third stage shows the 
 decay of the remaining radium C. 
 
 100, 
 
 15 
 
 30 
 
 45 60 
 
 Time in Minutes, 
 
 Fig. 10. 
 
 75 
 
 90 
 
 106 
 
 The curves for the same changes examined by 
 penetrating /& rays are given in figures 9 and 10 
 curve I. 
 
 The curve for a short exposure begins near zero. 
 This shows that radium A emits no ft rays. It then 
 
 62 
 
84 BEYOND THE ATOM [CH. 
 
 rises to a maximum in 36 minutes and afterwards 
 decays by an exponential law falling to half- value in 
 28 minutes. 
 
 With a long exposure not only will radium A have 
 been formed on the wire, but radium B and radium 
 C will have been produced from it. 
 
 Since (curve I, fig. 10) there is no sudden initial 
 drop, as in the a curves, we see again that the radium 
 A gives no rays. Curve II shows how the radium 
 C already existing when the wire is removed from the 
 emanation would decay, by a 28 minute law. Curve 
 III is the difference between curves I and II. It is 
 identical with the curve of fig. 9 for a short exposure, 
 rising to a maximum in 36 minutes and then decaying 
 by a 28 minute law. But this represents the activity 
 due to the fresh radium C produced from the radium 
 B. It follows that the change from B to C does not 
 itself emit any ft rays, but only gives rise to them 
 indirectly through the radium C which it produces. 
 We have already seen that the transformation of 
 radium B to radium C is not accompanied by a rays. 
 This change is therefore a ' ray less' one. It shows 
 that even a change which gives rise to no radiations 
 may be traced, if it is followed by active changes, by 
 the irregularities it introduces into their curves. 
 
 The results up to this point are summarized in the 
 curves of fig. 11. 
 
 A A represents the a ray activity of the rapidly 
 
VI] 
 
 A FAMILY TREE 
 
 85 
 
 decaying radium A. When this is subtracted from 
 the curve A + B 4- (7, there results the curve LL, 
 which is analysed into CO and BB as before. 
 The conviction of the truth of this analysis would 
 
 100 
 
 80 
 
 60 
 
 40 
 
 20 
 
 B+c 
 
 10 
 
 20 
 
 30 40 
 
 Time in Minutes, 
 
 Fig. 11. 
 
 50 
 
 60 
 
 70 
 
 be immensely strengthened, could we lay before the 
 reader the close agreement between the results 
 calculated from the mathematical equations repre- 
 senting the curves and the values experimentally 
 observed. 
 
86 BEYOND THE ATOM [CH. 
 
 Activity of Slow Change. 
 
 A similar patient investigation has been ap- 
 plied to the minute activity that remains when the 
 activity of rapid change has passed away. This 
 residual activity has also been resolved into three 
 stages : 
 
 (1) Radium D, derived from radium (7, is trans- 
 formed to half value, by another rayless change, in 
 about 40 years. It produces radium E. 
 
 (2) Radium E is transformed to half value in 
 about 5 days. It emits and 7 rays, and produces 
 radium F. 
 
 (3) Radium F is transformed to half-value in 
 143 days. It emits a rays. 
 
 For historical interest the original analysis as 
 given by Rutherford has been followed and the curves 
 are those published in his Bakerian lecture. Since 
 these initial experiments a large amount of work has 
 been carried out to examine in the greatest detail 
 this important series of changes. This has resulted 
 in some modifications, which do not however affect 
 the general theory. 
 
 For example, it has been found that radium B 
 gives out soft ft rays. These were not observed at 
 first because they were absorbed by the screens used 
 to absorb the a rays. Recently (Jan. 1912) it has been 
 
vi] A FAMILY TREE 87 
 
 found that radium B also emits some weak 7 rays. 
 The general analysis is not affected, but the periods 
 ascribed to B and C are interchanged, the a ray 
 product radium C having a period of 19*5 minutes 
 while radium B has the longer period of 26*8 
 minutes. 
 
 Those who have not had an opportunity of 
 following the theoretical analysis in detail, to whom 
 therefore it may seem a slender basis for accepting 
 the separate existence and properties of such sub- 
 stances as radium B and radium (7, admittedly present 
 only in infinitesimal quantity and in some cases not 
 even indicating their presence by radiations, may 
 perhaps be reassured to learn that it is now possible 
 to isolate radium B and C by heating a platinum 
 wire covered with the active deposit, and also by 
 direct chemical methods. Thus, if the active deposit 
 on a wire is dissolved in hydrochloric acid and a 
 nickel plate is dipped in the solution, radium C is 
 deposited on the plate and, when tested, proves to 
 have a period of 19*5 minutes. 
 
 Radium C emits the swiftest a rays of the series 
 and also very penetrating /3 and 7 rays, the ft rays 
 attaining the extraordinary speed of "998, all but two- 
 thousandths of the velocity of light. The explosion 
 of the radium C atom is unusually violent, and recent 
 results by Halm and Fajans have shown that it 
 appears to break up in two distinct ways ; for another 
 
88 BEYOND THE ATOM [CH. 
 
 substance, existing in very small amount, has been 
 discovered, which is half-transformed in 1*4 minutes 
 and emits /3 and 7 rays. Other instances are now 
 known where one product can break up in two 
 ways. Thus uranium breaks up into uranium X and 
 uranium Y, and thorium B into thorium d and 
 thorium O 2 . 
 
 With radium F the series of the descendants of 
 radium ends, for here the activity disappears. What 
 may happen next is matter for surmise. Meanwhile 
 its ancestors have been sought for with some 
 success. 
 
 Radium itself falls to half-value in about 2000 
 years. This is a long life indeed, but a mere drop in 
 the bucket compared with geological time. Had the 
 whole earth been pure radium originally, the amount 
 left now would have been utterly insignificant. 
 Radium must therefore have a parent, and it must be 
 one of the elements which have a heavier atom than 
 radium itself. Uranium seemed the most likely. 
 Its atomic weight is 238. It has now been proved 
 that uranium gives out two a particles in changing 
 to uranium X. That of uranium X is therefore 
 234. Radium is 226. But Ur X only gives out 
 ft rays. There is thus room for intermediate changes, 
 in two of which an a particle counting 4 must be 
 given out, between Ur X and radium. A search in 
 this direction was suggested by many indications. 
 
vi] A FAMILY TREE 89 
 
 Uranium was always found in minerals containing 
 radium. Again, it was remarkable that the greatest 
 a ray activity in pitchblendes was between four and 
 five times that of uranium. Now radium, in the 
 course of the changes we have just traced, goes 
 through at least four transformations in which an 
 a particle is expelled. In an old mineral a state of 
 equilibrium must have been reached and all these 
 changes must be going on together. We should 
 therefore expect about five times the activity of 
 uranium. Acting on these hints, several investigators 
 made a thorough search of all the known radiferous 
 minerals. Strutt and Boltwood examined a large 
 number taken from all parts of the world, and found 
 that the ratio of the radium emanation they contained 
 to the uranium present was almost exactly constant in 
 spite of great differences in composition and locality 
 of origin. This was a strong indication that uranium 
 and radium were causally connected. Accordingly 
 attempts were made to groiv radium from uranium. 
 Soddy examined the emanation produced from one 
 kilogram of uranium nitrate which had been carefully 
 freed from all trace of radium. Now calculation from 
 their rates of decay shows that the amount of radium 
 produced from one kilogram in a single day should 
 be easily detected. But Soddy's experiment showed 
 no result for two years, and then only a minute trace, 
 5*2 x 10~ u gm. per kgm. of uranium. 
 
90 BEYOND THE ATOM [CH. vi 
 
 This disappointing result shows that radium is not 
 directly produced from Ur X. But it is what would 
 be expected if there were an intermediate product of 
 very slow transformation, and therefore long life. 
 
 To understand this it must be remembered that in 
 a state of equilibrium the amount of each product in 
 the series transformed per second is the same. Hence 
 for a product of slow transformation, since only an 
 infinitesimal fraction of the amount existing is trans- 
 formed per second, that amount must be corre- 
 spondingly large. If, therefore, the uranium is cleared 
 of Ur X and all its products, before any radium can 
 again be detected, the uranium, itself transformed 
 very slowly it has a period of 6,000,000,000 years 
 must first accumulate the equilibrium amount of 
 Ur X, while this in turn has to accumulate the very 
 large equilibrium amount of the slow-transformation 
 parent of radium. 
 
 As a matter of fact Boltwood has recently separa- 
 ted the long sought substance from uranium minerals, 
 and named it Ionium. It is a close neighbour of 
 thorium and has an atomic weight of 230. It emits 
 an a particle and falls to half- value in about 200,000 
 years. Ionium can be initially freed from radium, 
 and then in the course of a month a considerable 
 amount of fresh radium, as tested by the emanation, 
 is found to have been produced in it ; moreover the 
 growth is proportional to the time. 
 
The Radio-active Families. 
 I. 
 
 Substance and 
 atomic weight 
 
 Period of fall to 
 half-value 
 
 ! 
 
 Radiations 
 emitted 
 
 Range of a particles 
 in air at 15 C. 
 centimetres 
 v. p. 138 
 
 Uranium 1 238-5 
 
 6 x 10" years 
 
 a 
 
 2-50 
 
 Uranium a 
 
 2 x 10 years 
 
 a 
 
 2-90 
 
 Uranium X 230-5 
 | 
 Ionium 230 '5? 
 
 i 
 
 24-6 days 
 200,000 years 
 
 fiy 
 
 a 
 
 3-0 
 
 Kadium 226-5 
 
 i 
 
 2000 years 
 
 ft 
 
 3-30 
 
 Emanation 
 
 i 
 
 3-85 days 
 
 a 
 
 4-16 
 
 Kadium A 
 
 3 utiiis. 
 
 a 
 
 4-75 
 
 Radium B 
 Badium C 
 
 26*8 mins. 
 19-5 mins. 
 
 ft 
 apy 
 
 6-94 
 
 Badium Dj 
 
 1-4 min. 
 
 fty 
 
 
 Badium Z>2 
 Badium E l 
 Badium # 2 
 
 15 years 
 4-8 days 
 
 ft 
 
 
 Badium F 
 (Polonium) 
 
 140 days 
 
 a 
 
 3-77 
 
 The break-up of radium C is abnormal. Geiger and Marsden 
 find that in this case, as also for thorium C and the emanations of 
 thorium and actinium, the scintillations from the a particles occur 
 in pairs. Either, then, the atoms break up in two different ways, 
 or there are two successive products, one of them very short-lived. 
 
IL 
 
 Substance 
 
 Period of fall to 
 half-value 
 
 Radiations 
 emitted 
 
 Range of a particles 
 in air. Centimetres 
 v. p. 138 
 
 Thorium 232-5 
 
 3-10 10 years 
 
 a 
 
 2-72 
 
 I 
 
 
 
 
 Mesothoriumj 
 
 5-5 years 
 
 rayless 
 
 
 Mesothorium 2 
 
 6-2 hours 
 
 Pv 
 
 
 | 
 
 
 
 
 Radio-thorium 
 
 2 years 
 
 a 
 
 3-87 
 
 Thorium X 
 
 3-70 days 
 
 a/3 
 
 4-30 
 
 I 
 
 
 
 
 Emanation 
 
 54 sees. 
 
 a 
 
 5-00 
 
 Thorium A 
 
 1 
 
 14 sec. 
 
 a 
 
 5'70 
 
 Thorium B 
 i 
 
 10-6 hours 
 
 f 
 
 
 rc 
 Thorium [^ 
 
 60 mins. 
 
 a 
 a 
 
 4-80 
 8-60 
 
 1 
 
 
 
 
 Thorium D 
 
 3 mins. 
 
 7 
 
 
 Actinium 
 1 
 
 
 
 
 Radio-actinium 
 
 19-5 days 
 
 a/3 
 
 4-60 
 
 1 
 
 
 
 
 Actinium X 
 
 10-5 days 
 
 a 
 
 4-40 
 
 I 
 
 
 
 
 Emanation 
 
 3-9 seconds 
 
 a 
 
 5-70 
 
 I 
 
 
 
 
 Actinium A 
 
 36-1 mins. 
 
 Q 
 
 
 \ 
 
 
 
 
 Actinium B 
 
 2 '15 mins. 
 
 a 
 
 6-50 
 
 \ 
 
 
 
 
 Actinium C 
 
 5-1 mins. 
 
 07 
 
 5-40 
 
 For the two emanations the latter view is confirmed by a study of 
 
CH. vi] A FAMILY TREE 93 
 
 The complete family tree of the uranium-radium 
 family as far as it is known at present is given in 
 tabular form, and similar tables are added for 
 thorium and actinium. 
 
 The second column gives for each product the 
 time it takes to fall to half- value. This is a constant 
 by which the product can be identified. 
 
 By means of these constants it has been shown 
 
 the ranges of the a particles emitted. From Rutherford's law 
 connecting the life of a product with the range of its a particle, it 
 appears that one of the products of the thorium emanation has a 
 period 0'14 sec., and the corresponding product of the actinium 
 emanation has the minute period, *002 sec. ! A new a ray product 
 with this period must therefore be inserted in the table ahead of 
 actinium A. 
 
 The existence of these short-lived products may be shown by a 
 pretty experiment. An endless charged wire is run on insulating 
 pulleys at high speed. In passing through a box it is exposed to 
 actinium or thorium emanation, and immediately outside it runs 
 along a zinc-sulphide screen, which is seen to be lighted up for some 
 distance from the box. From the speed of the wire the life of the 
 a ray product deposited on it by the emanation may be determined. 
 
 For radium C and thorium C (or, rather, thorium J5) it would 
 seem that there is really a double break-up, a branch being thrown 
 off from the main line of descent. Some recent work by Marsden 
 and Darwin tends to show that one-third of the atoms of Th B break 
 up into Th C l , which emits a rays of range 4-8 and produces the 
 fiy ray product Th D ; while two-thirds break up into Th C 2 , a /S ray 
 product whose subsequent history is not yet known. Perhaps this is 
 because the product it ends in emits the fastest a particles known, 
 with a range 8-6, so that by Rutherford's law its period should have 
 the incredibly minute value 10~ 12 sec. 
 
94 BEYOND THE ATOM [OH. 
 
 that the active principle in polonium is really the 
 seventh disintegration product of radium, radium F ; 
 and the radio-lead owes its activity to radium D and 
 its products. These discoveries are indicated in the 
 table. 
 
 The atomic weight falls by 4 every time an a 
 particle is ejected ; but is not diminished appreciably 
 by the expulsion of an electron, or a 7 ray. The 
 series starts with uranium 238 ; this is now known to 
 give out two a particles, with the formation of a 
 second substance, uranium 2, before uranium X 230 
 is reached. Ur X only gives out /3 and 7 rays, 
 producing ionium, still 230, and then by the expulsion 
 of another a particle radium is reached with atomic 
 weight 226. 
 
 What becomes of radium F ? Since it gives out 
 an a particle, the next product should have an atomic 
 weight 206, which is near that of lead. Accordingly 
 Boltwood early suggested that the final result, when 
 the series of active changes is accomplished, should 
 be lead, an idea supported by much evidence as to 
 the proportion of lead found in the radiferous 
 minerals, similar to that which suggested the descent 
 of radium from uranium. 
 
 The final product of the thorium series is not so 
 certain. Six a particles are given out, so that its 
 atomic weight should be 232*5 24=208*5, which 
 suggests bismuth, but the evidence from the minerals 
 
vi] A FAMILY TREE 95 
 
 does not in this case fall into line. This may be due 
 to the double break-up at thorium (7, leading perhaps 
 to two end-products. 
 
 Nothing is known of the atomic weights of the 
 actinium series. There are indications that this 
 series is a branch from the uranium-radium family, 
 possibly taking its rise at the abnormal break-up 
 of radium C. 
 
 The Disintegration Theory has received (Proc. 
 Roy. Soc. Jan. 12, 1911) a beautiful confirmation. 
 Dr Gray and Sir William Ramsay have succeeded 
 in directly determining the molecular weight of the 
 radium emanation, by weighing it, and measuring 
 its volume, temperature and pressure. The total 
 volume of emanation available was less than one- 
 tenth of a cubic millimetre. Yet not only was 
 this mere pin's point of gas weighed, but elaborate 
 corrections were made for its decay during the 
 experiment, for the amount of helium occluded in the 
 tube, and other minute sources of error. The results 
 of five experiments range between 227 and 218 with 
 a mean of 223. Since the equivalent of radium is now 
 established at 226*4, that of the emanation, resulting 
 from the loss of one a particle, should theoretically 
 be 226'4 4 = 222*4, an extraordinarily close agree- 
 ment. 
 
 The work of Gray and Ramsay was only ren- 
 dered possible by an achievement in the southern 
 
96 BEYOND THE ATOM [OH. 
 
 hemisphere, which well illustrates the gain of power to 
 science through the world-wide co-operation of disin- 
 terested investigators. Dr Steele and Mr Grant of the 
 University of Melbourne have recently constructed a 
 balance with a girder beam made of fine silica rods, 
 and devised a method of weighing by displacement 
 with it which allows the estimation of one quarter of 
 a millionth of a milligram. The limit of sensibility 
 of the finest assay balance has hitherto been one two- 
 hundredth of a milligram. This marvellous increase 
 in delicacy is no more than was required for Gray and 
 Ramsay's experiment, for the total quantity of emana- 
 tion dealt with weighed only about 700 millionths of 
 a milligram. The credit of a wonderful feat of mani- 
 pulation is thus shared between two hemispheres. 
 
 CHAPTER VII 
 
 VERIFICATIONS AND RESULTS 
 
 THERE are four stages in the establishment of any 
 natural law. First, observation, the recognition of 
 some new fact or relation. This is usually the work 
 of genius. It requires insight to descry amidst the 
 familiar what has always been there but hitherto has 
 escaped notice. Some discoveries appear accidental, 
 but these accidents only happen to those with seeing 
 
vn] VERIFICATIONS AND RESULTS 97 
 
 eyes and imagination to grasp the significance of what 
 the dull and incurious would pass as unimportant. 
 Then comes experiment, by which the new facts are 
 made sure and disentangled from what is irrelevant. 
 Next, the consequences of the new view are traced 
 by deduction in detail. Lastly comes verification, 
 when those consequences are compared with further 
 experimental results. 
 
 The Disintegration Theory described in the last 
 two chapters effects so profound a revolution in 
 accepted ideas that it will be well to collect a few 
 leading verifications. But it must be borne in mind 
 that these are only samples of a vast mass of results 
 which, as experimental methods improve, are found 
 in increasing agreement with calculation. 
 
 1. Volume of the Emanation. 
 
 Quite early, in the course of an investigation of 
 the charge carried by the a particles, Rutherford 
 calculated the amount of emanation that should be 
 obtained from one gram of radium in radio-active 
 equilibrium. Apart from the very slow residual 
 changes, which for this purpose may be neglected, 
 there are then four sets of a particles being emitted, 
 from the radium, the emanation, radium A, and 
 radium C. When the radium is freed from the 
 emanation and subsequent products, it still retains 
 
 c. 7 
 
98 BEYOND THE ATOM [CH. 
 
 its own a activity which is one quarter of the total 
 This is called its minimum activity. The minute 
 current carried by the a particles from a known 
 weight of radium bromide at this minimum activity 
 was measured on a Dolezalek electrometer. If each 
 a particle carries the ionic charge, then by division 
 into the total charge carried it appears that the 
 number of a particles emitted per second is 3*6 x 10 10 *, 
 Le. thirty-six thousand million. For each a particle 
 emitted there is a corresponding atom of emanation 
 produced. Now the rate of decay of the emanation 
 is known from its curve. In fact 1 particle in every 
 463,000 breaks up per second. But in equilibrium 
 the number of particles breaking up must be equal 
 to the number of new particles produced. Thus 
 3-6 x 10 10 must be one 463,000th part of those 
 existing at any moment. The number of molecules 
 of emanation to be expected from one gram of radium 
 bromide, when the whole amount stored in it is 
 disengaged by solution, should thus be 
 
 463,000 x 3'6 x 10 10 , 
 
 i.e. 17 x 10 16 . The fraction of a cubic centimetre that 
 these would occupy at ordinary temperature and 
 pressure is obtained by dividing by Avogadro's con- 
 stant, the number of molecules of any gas in a cubic 
 centimetre. This is 3*6 x 10 19 . The volume of the 
 
 * But see p. 102. 
 
vii] VERIFICATIONS AND RESULTS 99 
 
 emanation from one gram of radium bromide is thus 
 "46 x 10~ 3 c.c., Le. *46 cubic millimetres. Allowing for 
 the chemical composition of the bromide (RaBr 2 ) 
 this shows that for pure radium the amount of 
 emanation would be '82 cubic millimetre per gram. 
 When Ramsay and Soddy at last succeeded in 
 isolating and measuring the minute bubble of emana- 
 tion from 60 milligrams of radium bromide, they con- 
 cluded that a gram of pure radium should produce 
 about one cubic millimetre of emanation, a result 
 sufficiently concordant with calculation when the 
 extreme difficulty of the experiment is remembered. 
 Recent investigations (1908) have, as we shall see, 
 brought about a much closer agreement. 
 
 2. Counting the Atoms. 
 
 Since on the assumption that an a particle carries 
 the ionic charge, a gram of radium bromide emits at 
 minimum activity thirty-six thousand million a 
 particles per second, each representing the break 
 up of an atom, it would seem a most hopeless quest 
 to detect the effect of a single a particle and so 
 count the atoms destroyed. Yet this is what has 
 been accomplished. 
 
 The action of individual a particles was first made 
 visible in the little instrument invented by Sir William 
 Orookes, the spinthariscope. A needle dipped for a 
 
 72 
 
100 BEYOND THE ATOM [OH. 
 
 moment in a weak radium solution is placed close in 
 front of a small zinc sulphide screen, and examined 
 by a lens. When the eye has been made sensitive by 
 remaining for some minutes in the dark, the surface 
 of the screen is seen to be twinkling incessantly with 
 thousands of points of greenish light. These are due 
 to the impact of a particles discharged by the radium, 
 but the exact nature of the process, and whether each 
 sparkl e indicated a single a particle, were questions 
 it was not easy to determine. 
 
 In 1908 Rutherford and Geiger devised an electrical 
 method of counting the particles. An a particle ex- 
 pelled from radium produces more than 100,000 ions in 
 air before it is stopped. Calculation from this datum 
 shows that while the detection of a single a particle 
 by its ionizing effect should be just within the range 
 of the most refined experiment, it would be beset 
 with grave difficulties. Now Townsend had shown that 
 when ions are produced in an electric field their speed 
 may be increased till, first, the negative ions and, later, 
 when the potential gradient approaches the sparking 
 limit, the positive ions produce many more ions by 
 collision with the neutral atoms of the gas. It 
 occurred to Rutherford to employ this method of 
 magnifying the ionizing effect of a single a particle. 
 
 The source of a particles was a small disc exposed 
 for some hours to radium emanation. This was 
 carried on the end of a small iron cylinder which 
 
vii] VERIFICATIONS AND RESULTS 501 
 
 could slide along a long glass tube and be moved to 
 any position by means of a magnet without opening 
 the tube. Fifteen minutes after removal from the 
 emanation the radiation from the disc was practically 
 reduced to that from radium C. At the other end of 
 the tube, some four metres distant, was a small open- 
 ing, covered with a thin mica plate, into the exhausted 
 testing chamber, which consisted of a cylinder with a 
 central insulated wire. The voltage applied between 
 the wire and the cylinder was adjusted so that the 
 ionizing effect of a constant source of 7 rays was 
 multiplied several thousand times. 
 
 The a particles are shot out in all directions, so 
 that a few of them enter the chamber through the 
 distant narrow aperture. It was found possible to 
 arrange that from two to five per minute entered the 
 vessel, each one signalising its arrival by an un- 
 mistakeable throw of the needle of an electrometer 
 connected in the usual way with the testing vessel. 
 The results were admirably consistent, and it was 
 then easy to calculate from the known size of aperture 
 and distance from the source what must be the total 
 number of particles emitted in all directions. In 
 three experiments when the source was distant three 
 metres and a half from the aperture, only one particle 
 out of 125,000,000 emitted would get into the vessel. 
 
 Making all corrections, Rutherford and Geiger 
 concluded that the number of a particles emitted 
 
108 BliYOND THE ATOM [OH. 
 
 from one gram of radium itself per second is 
 3-4 x 10 10 , instead of 3'6 x 10 10 as assumed above. If 
 the radium is in equilibrium with its products, since 
 the emanation, radium A, and radium C each give 
 out an a particle for every one given out by the 
 radium, the number will be four times as great. 
 
 Experiments were then carried out in which the 
 number of scintillations were counted under a micro- 
 scope. The close concordance of the results proves 
 that in the spinthariscope each scintillation corre- 
 sponds to the impact of one a particle. 
 
 The visible detection, by two methods, of the 
 effect of a single particle of atomic size is certainly 
 one of the most remarkable and least expected 
 triumphs of experimental skill. 
 
 3. Proof that a particles are Helium atoms. 
 
 The determination of the number of a particles 
 emitted, by two methods of directly counting them, 
 enabled Rutherford and Geiger to settle the vexed 
 question whether these particles are really helium 
 atoms, as had early been suggested by Rutherford. 
 In 1908 they repeated the experiments for measuring 
 the total charge carried by the a particles from a 
 definite amount of radium, and showed from a series of 
 very consistent results that the charge of an a particle 
 is 9*3 x 10~ 10 electrostatic units. 
 
vii] VERIFICATIONS AND RESULTS 103 
 
 The ionic charge, as we have seen, had been found 
 by J. J. Thomson to be 3'4 x KT 10 . Other deter- 
 minations by different methods were, by Townsend 
 3 x 10- 10 ; H. A. Wilson 3'1 x 1Q- 10 ; Millikan 
 4'06 x 10- 10 and more lately 4'9 x lO" 10 . Rutherford 
 and Geiger calculated the value 4*1 x 10~ 10 by an 
 ingenious application of the new method of counting 
 the a particles to Boltwood's value of the life of 
 radium, which however was too low. Knowing the 
 number of atoms of radium which break up per second 
 and from its decay curve the fraction which this is of 
 the whole, we obtain the number of atoms in a gram 
 of radium ; thence, from the atomic weight of radium, 
 the number in a gram of hydrogen ; and finally from 
 the hydrogen liberated by a definite amount of electri- 
 city in electrolysis, the charge carried by each atom. 
 
 But then, what meaning is to be attached to the 
 value 9*3 x 10~ 10 for the charge of an a particle? It 
 must be some multiple of the unitary charge carried 
 by an ion. Is it twice, or three times as great? 
 Rutherford has given very strong reasons for believing 
 that the early determinations of the ionic charge 
 were all too low, one source of error being that no 
 allowance was made for evaporation from the drops 
 during the settlement of the clouds used for counting 
 the molecules (v. Chapter II). In that case the 
 a particle must carry double the ionic charge, and 
 the latter must be taken to be not 3*4 x 10~ 10 but 
 
104 BEYOND THE ATOM [OH. 
 
 4*65 x 10~ 10 . This would agree with recent values 
 obtained by Planck from the theory of radiation and 
 by Regener, who introduced a beautiful modification 
 of Crookes' spinthariscope method by counting the 
 scintillations produced by the a particles from polon- 
 ium upon a small diamond plate. 
 
 Now the ratio e/m for an a particle is about 
 half that for the hydrogen atom in electrolysis. If 
 its charge is double, then the mass must be four 
 times that of a hydrogen atom. To be precise, it 
 comes out 3*84, while the atomic weight of helium 
 is 3*96. Allowing for experimental errors, it seems 
 reasonable to conclude that an a particle is an atom 
 of helium carrying a double ionic charge. 
 
 The first attempts to show experimentally that 
 helium is produced from the radium emanation were 
 made by Ramsay and Soddy in 1903, following on 
 Rutherford's suggestion in 1902. Dewar and Curie 
 made similar experiments afterwards. In each case 
 a quantity of emanation was enclosed in a small 
 spectrum tube and the spectrum examined day by day. 
 The helium lines gradually appeared. 
 
 These experiments were of extreme difficulty 
 and it might, as indeed it was, still be argued that 
 the helium had nothing to do with the a particles 
 but was a by-product of the explosion of the radium 
 atom like the emanation itself, one atom of helium 
 being produced and one of emanation for every a 
 
vii] VERIFICATIONS AND RESULTS 105 
 
 particle expelled. But in 1909 Rutherford and Royds 
 devised a more direct demonstration. They succeeded 
 in blowing very thin glass bulbs, less than one 
 hundredth of a millimetre (^-gV^h of an inch) in thick- 
 ness, which would allow the a particles to pass through 
 the glass and yet stand atmospheric pressure. One of 
 these bulbs filled with radium emanation was enclosed 
 in an outer glass tube which was exhausted, and the 
 spectrum obtained when a current was passed through 
 the outer tube was examined. For twenty-four hours 
 no sign of helium appeared. After four days the 
 helium lines in the yellow and green were seen. In 
 six days the complete spectrum of helium was obtained. 
 Since nothing could gain access to the outer tube 
 except the a particles fired through the walls of the 
 thin glass bulb, it was impossible to avoid the con- 
 clusion that the a particles are themselves atoms of 
 helium. As a check experiment, the inner bulb was 
 then filled with helium under pressure, but no sign of 
 helium appeared in the outer tube. 
 
 4. Rate of Production of Helium. 
 
 When radium has stood long enough to be in 
 equilibrium with its products there are four products 
 giving out a particles : radium, the emanation, and 
 radium A and C. Each of these gives out, by the 
 
106 BEYOND THE ATOM [OH. 
 
 counting experiments, 3'4 x 10 10 particles per second. 
 All four together give out in one year 
 
 4 x 3-4 x 10 10 x 365 x 24 x 60 x 60 particles per gram. 
 
 Dividing by 272 x 10 19 , the revised value of the 
 number of molecules in a cubic centimetre of any gas, 
 we find the yearly product of helium from a gram of 
 radium to be *158 c.c., or 158 cubic millimetres. 
 Recent experiments, by Dewar in 1910, gave a yield 
 of 170 cub. mm.; and still more recently Boltwood 
 and Rutherford have obtained the value 156 c.c. This, 
 as the authors say, is Disagreeably' close to the 
 theoretical number, but experiments by two different 
 methods gave the same result. 
 
 Similarly, if the latest values are employed, the 
 volume of emanation from a gram of radium is found 
 to be '585 cubic mm. instead of *82; and this is in 
 agreement with the best experimental measures, viz. 
 61 cub. mrn. (Rutherford, 1908), "58 mm. (Debierne, 
 1909), '601 mm. (Ramsay and Gray). 
 
 For comparison with the atomic constants given 
 in Chapters I and II we may here collect the 
 results based on Rutherford's direct counting ex- 
 periments. 
 
 Number of a particles per second 
 from 1 gram of radium at mini- 
 mum activity . . . . 3'4 x 10 10 
 Charge of an a particle . . . 9'3 x 10~ 10 E.S. 
 
vii] VERIFICATIONS AND RESULTS 107 
 
 Ionic charge (half above), i.e. charge 
 
 on hydrogen atom . . . 4*65 x 10 10 
 Consequent number of hydrogen 
 
 atoms in cubic centimetre . . 272 x 10 19 
 Mass of hydrogen atom . . . 1*61 x 10~ 24 gm. 
 Volume of emanation per gram of 
 
 radium. . . . . '585 cub. mm. 
 
 Volume of helium per gram per 
 
 year 158cub.mms. 
 
 5. Recoil of the Atoms 1 . 
 
 If it is true that the atoms of any product, in the 
 act of transformation to the succeeding product, fire 
 out an a particle with high speed, then the atoms 
 of the new product so formed should recoil with 
 the same momentum as that communicated to the 
 a particle, exactly as a gun recoils on discharging 
 its projectile. This is actually found to be the case. 
 A metal plate set near a radio-active substance is 
 bombarded by these recoil-atoms, and collects the 
 succeeding product in a pure state. This method 
 is now freely employed to separate out some products, 
 such as radium B, whose existence could only be 
 inferred from the curves, and by means of it some 
 new products have been detected. Moreover, the 
 recoil-atoms can be deflected in a magnetic field. 
 Their deflection is half that of the a particle, and 
 
 1 Kuss and Makower, Proc. Roy. Soc. 1909. 
 
108 BEYOND THE ATOM [CH. 
 
 therefore they carry half the charge of an a particle, 
 i.e. the ionic charge. This method also affords the 
 only means of determining the atomic weights of 
 many products which exist in very minute quantity. 
 
 6. Emission of Energy by Radium. 
 
 Although it was certain that energy was being con- 
 stantly emitted by the radio-active substances, it 
 came as a startling surprise when P. Curie and Laborde 
 in 1903 announced that radium compounds main- 
 tained themselves permanently at a temperature two 
 or three degrees higher than their surroundings. 
 Measurements by the Curies, by Rutherford and 
 Barnes, and later by Callendar, Angstrom, Schweidler 
 and Hess, the latter using the large quantity of radium 
 separated by the Austrian Government, show that 
 1 gm. of radium emits heat at the rate of 132 gram- 
 calories per hour, enough to melt 1*6 times its own 
 weight of ice per hour, and that three quarters of 
 the heat emission is due to the emanation and its 
 products. This is what should be expected. From 
 their masses and velocities it is easy to show that 
 the energy of an a particle is about one hundred 
 times that of the average & particle. The heating 
 effect is thus due chiefly to the a particles, and 
 the emanation, with its products radium A and (7, 
 emits three a particles to each one emitted by radium 
 
vn] VERIFICATIONS AND RESULTS 109 
 
 itself. As a further verification, the heat calculated 
 from the number of a particles emitted and their 
 known energy is in fair agreement with the value 
 found experimentally. This gives a total disengage- 
 ment of heat from one gram of radium, before it dis- 
 appears, about one million times as great as is evolved 
 in the formation of one gram of water. One pound 
 of the emanation would at its maximum intensity 
 radiate energy at the rate of about 10,000 horse power. 
 This large emission of energy from the radio-active 
 bodies throws an interesting light on two questions 
 which have long been the subject of controversy, the 
 age of the earth and the source of the heat of the sun. 
 
 7. The Age of the Earth. 
 
 In descending a mine it is found that the tempera- 
 ture rises about one degree Fahrenheit for every 50 
 or 60 feet of descent. Lord Kelvin showed long ago 
 that if the earth be regarded as a body cooling freely 
 in space without internal supply of heat, this tempera- 
 ture gradient indicates that about 100,000,000 years 
 ago the earth must have been a molten mass. This 
 period, long as it seems, is quite insufficient for geo- 
 logists to account for the deposition of the observed 
 strata, and for biologists, to allow for the development 
 of known types from primordial life. Hence a some- 
 what embittered controversy. 
 
110 BEYOND THE ATOM [CH. 
 
 The discovery that radium and its emanation, not 
 to speak of the other radio-active elements, are widely 
 distributed, though in minute proportion, throughout 
 the atmosphere, the springs, the rocks and soils of the 
 earth's crust removes the difficulty. Henceforth the 
 physicists can make the geologists and biologists 
 happy with a blank cheque on eternity, so far as this 
 argument is concerned. The present loss of heat from 
 the earth is about 2-2 x 10~ 7 calories per c.c. per year 
 which could be supplied indefinitely by the presence 
 of 2 x 10~ 13 gram of radium per c.c. Now Strutt finds 
 there is on the average 4 x 10~ 12 gram of radium per c.c., 
 i.e. about twenty times more than is required. There 
 is thus no reason why the present state of things should 
 not have been maintained for far longer periods than 
 100,000,000 years. On the contrary, it is necessary to 
 suppose that the percentage of radium found at the 
 surface is confined to a crust about forty miles thick, 
 and that the interior is free from radium. This would 
 indicate a temperature of about 1500 C. for the core, 
 a value not inconsistent with the temperature of the 
 lava from Mount Etna, 1060 C. The great heat met 
 with occasionally, as in the boring of the Simplon 
 tunnel, is also accounted for, since in regions rich in 
 activity the temperature gradient is abnormally high. 
 
vn] VERIFICATIONS AND RESULTS 111 
 
 8. The Age of Rocks. 
 
 But the discoveries of the last ten years have done 
 more than remove a difficulty for the geologists. 
 They supply a means of directly estimating the age of 
 radiferous rocks and, by implication, of the strata in 
 which they are found, through the estimation of their 
 content of helium. This element has a strange story. 
 Its existence was first inferred from certain spectral 
 lines only seen in a total eclipse of the sun. Hence 
 the name. Then it was most unexpectedly disengaged 
 from cleveite, a heavy mineral. It is only found in 
 radio-active rocks and is occluded in them where it is 
 produced, as uranium degrades into radium and its 
 successive products. If therefore the annual pro- 
 duction of helium due to the uranium present be 
 divided into the total amount of helium found, some 
 hint should be obtained of the number of years during 
 which it has been storing up. 
 
 Thus Fergusonite contains 7 per cent of uranium 
 and 1'8 cubic c.c. of helium per gram. Now one gram 
 of uranium, with its products, gives out a particles to 
 the extent of 1*08 x 10~ 7 c.c. of helium per year. 
 Thus the age of Fergusonite should be 
 
 Le. about 240 million years. 
 
 Thorianite contains the unusual amount of 9'5 c.c. 
 
112 BEYOND THE ATOM [CH. 
 
 of helium per gram. In this case Strutt has estimated 
 the age by directly measuring the annual production 
 and dividing into the total found 1 . 
 
 Another point which may interest biologists as 
 well as geologists arises from the discovery of radium 
 emanation in mineral springs. It may possibly be 
 the secret of their medicinal effects, which in many 
 cases cannot be traced to any special chemical com- 
 position, and disappear when the waters are bottled 
 and sent to a distance. Now the emanation certainly 
 has a physiological effect, and would have passed 
 away in less than a month. 
 
 9. Age of the Sun. 
 
 The discovery of radium has not brought quite 
 the same relief to astronomers with regard to the 
 question of the source of the sun's heat. Lord Kelvin 
 in a famous passage discussing the theory of gravi- 
 tational contraction from a nebula, the only one which 
 can at all account for the sun's enormous output, 
 concludes that it is ' on the whole probable that the 
 sun has not illuminated the earth for 100,000,000 
 
 1 The age may also be estimated by the content of lead. Helium, 
 since part of it may escape, gives a minimum value ; lead, which does 
 not escape and may not all have been formed in situ, gives a maximum, 
 from two to four times the value from helium. Thus for the Carboni- 
 ferous Limestones (Haematite, Forest of Dean) Strutt assigns an age 
 of 150 million years from helium, and 340 million from the lead- 
 content. 
 
vn] VERIFICATIONS AND RESULTS 113 
 
 years and almost certain that he has not done so for 
 500,000,000 years. As for the future we may say, 
 with equal certainty, that inhabitants of the earth 
 cannot continue to enjoy the light and heat essential 
 to their life for many million years longer, unless 
 sources now unknown to us are prepared in the great 
 storehouses of creation.' 
 
 Had Kelvin had at his disposal Langley's instead 
 of Pouillet's measurement of the solar constant of 
 radiation, his hundred million would have been 
 reduced to sixty ; and Sir George Darwin has given 
 reasons why this again should be brought down to 
 twelve. 
 
 On the discovery of radium the cautious passage 
 we have italicised seemed almost prophetic. It is 
 true that no 7 rays from radium have been detected 
 in sunlight; but this is not surprising since the 
 atmosphere which shields the earth is equivalent to 
 a thickness of thirty inches of mercury. Nor have 
 the radium lines yet been observed in the solar 
 spectrum. But this is not the difficulty. To account 
 for the sun's output of 430 calories per hour per 
 cubic metre, would require 3*2 gms. of radium per 
 cubic metre, since each gram gives 132 calories. Now 
 the sun's density is 1'44, so that he should contain 
 1'8 x 10~ parts by weight of radium. This is five 
 or six times too great for radium in equilibrium 
 with uranium. 
 
 c. 
 
114 BEYOND THE ATOM [CH. 
 
 Radium cannot, as was at first supposed, supply 
 the whole solar output. But it may very well be a 
 large contributor ; and it is possible that at the high 
 temperatures existing in the sun radio-active processes 
 may go forward at a greater rate than at the temper- 
 atures available on earth, while other instances of 
 transformations in which atomic energy is unlocked, 
 perhaps without the accompaniment of radio-active 
 phenomena, may be taking place. 
 
 10. Cosmical Effects. 
 
 The aptness of electrons and positive ions to form 
 nuclei for the condensation of vapours has been 
 applied by Arrhenius to explain a number of pheno- 
 mena which have long puzzled astronomers. 
 
 1. Comets' Tails. These are produced as the 
 comet approaches the sun : shoot out to enormous 
 distances, always pointing away from the sun what- 
 ever the direction of the comet's motion, so that 
 in departing the tail precedes the head; and, as 
 the comet sweeps round the sun at perihelion, are 
 1 brandished,' as Herschel says, through half a circle 
 without injury, inconceivable behaviour for such an 
 enormously extended appendage of matter which, 
 from its transparency to the stars and lack of 
 attraction on the planets, must be far more rarefied 
 than our best earthly vacuums. 
 
vii] VERIFICATIONS AND RESULTS 115 
 
 Arrhenius suggests that the hydro-carbon vapours 
 roasted out of the comets' materials by the tremen- 
 dous heat of the sun condense into droplets on the 
 electrons which the sun, as a very hot body, must 
 discharge into space, and are then driven backwards 
 by the pressure of the sun's light. This pressure, 
 predicted and calculated by Maxwell in 1867, is so 
 minute on surfaces of ordinary size, that it was only 
 discovered experimentally in 1904. But since the 
 surface exposed to pressure falls off as the square 
 of a particle's diameter, while the solid contents 
 subject to gravitation fall off as the cube, as we 
 proceed to smaller and smaller particles there must 
 come a time when the light-pressure will balance the 
 weight. For liquid droplets this 'critical' value is 
 about one micron. For still smaller diameters the 
 light-pressure will exceed the solar gravitation, and 
 may be as much as eighteen times as great. Here 
 at last is the ' violent repulsive force from the sun/ 
 the ' negative gravity ' so long sought in vain. Such 
 a force would make light of projecting a tail 20,000,000 
 leagues in two days, as Newton observed in the case 
 of the great comet of 1680. It is not, then, the same 
 tail which is brandished, but the tail is constantly 
 being re-formed, and always in the direction away 
 from the sun. 
 
 2. The Aurora. The vapours ejected from the 
 sun must condense and be driven off in the same 
 
 82 
 
116 BEYOND THE ATOM [OH. 
 
 way, with speeds which would carry them beyond 
 the earth's orbit in a few hours. On this Arrhenius 
 founds plausible explanations of the Corona and 
 Prominences observed during eclipses, but especially 
 of the Aurora. The flights of droplets, each carrying 
 a negatively charged ion, arriving over the equator 
 are caught by the lines of force of the earth's magnetic 
 field, which arch from pole to pole, and, by well- 
 known electrical laws, are obliged to follow them 
 in spirals towards the poles. In their passage 
 through the rarefied upper air they give rise to the 
 shifting lights which may be mimicked on a small 
 scale by the discharge of similar particles from the 
 kathode of a vacuum tube. The writer has suggested 
 that the strange and beautiful ' curtain' auroras, 
 taking the form of long fringes of coloured light, 
 hanging in vertical lines and sharply terminated 
 at the lower edge, may be due to similar discharges 
 of a particles, which would have a long range in the 
 rare atmosphere at the upper levels, but be suddenly 
 extinguished in the end. 
 
 The theory solves another difficulty. It had 
 never been understood how disturbances on the sun, 
 whatever their violence, could directly aflfect the 
 magnetic state of the earth. Yet when records of 
 the number of Sun-spots, Magnetic Storms and 
 Aurorae are plotted over a long interval of years, the 
 three curves follow each other closely, showing the 
 
vin] REALITY OF MOLECULES 117 
 
 same slow secular changes, the well-known 'eleven 
 year' period, the annual variation, and even two 
 smaller monthly variations ; so that there is an 
 undoubted connection between them. It is obvious 
 on Arrhenius' view that the flights of electrons are 
 the equivalent of negative currents and must produce 
 magnetic disturbances on earth ; and he is able to 
 account in detail for the variations on the very 
 reasonable hypothesis that the sun-spots, as the seat 
 of violent eruptive storms, are the most prolific 
 source of electrons. 
 
 CHAPTER VIII 
 
 THE OBJECTIVE REALITY OF MOLECULES 
 
 A STUDENT of the tables on pp. 10, 106 may well be 
 staggered by the inconceivable minuteness of some 
 of the quantities in which he is called upon to believe, 
 and perhaps ask himself whether in the kinetic theory 
 of matter which leads to such results we are proceed- 
 ing upon any basis of reality, or whether it be not 
 after all merely an ingenious hypothesis. It is true 
 that by means of it we can co-ordinate the experi- 
 mental laws discovered in many branches of physics ; 
 and the conclusions drawn from widely different 
 physical phenomena as to the size, mass and number 
 
118 BEYOND THE ATOM [OH. 
 
 of the atoms, and the charge of an electron, show 
 a remarkable concordance, at least as to the order 
 of magnitude. But the largest molecule is far beyond 
 direct sense perception ; the atom is a long stage 
 beyond the molecule ; and the electron is another 
 thousandfold stage beyond the atom. Moreover the 
 violent agitation in which these supposed entities 
 must be conceived to exist, not only in gases but 
 even in liquids and solids, appears to contradict the 
 first direct evidence of the senses. Newton's warning 
 ' Hypotheses non fingo ' seems in place at this point, 
 and a verification which should give to the molecule 
 and the atom the character of ' verae causae ' would 
 be welcome indeed. 
 
 Most opportunely such a verification is to hand 
 in Perrin's recent study (1908-10) of the so-called 
 Brownian movement, which renders visible in the 
 microscope not, indeed, the molecules themselves, 
 but a direct effect of their perpetual movement in 
 a liquid. The phenomenon was first noticed in 1827 
 by the English botanist Robert Brown, shortly after 
 the discovery of the achromatic objective. Brown, 
 while engaged in examining pollen grains, observed 
 that very minute material particles suspended in 
 a liquid were never at rest, however long the 
 liquid had been standing, but subject to incessant, 
 irregular movements, so that they described paths, not 
 curvilinear, but made up of short disjointed straight- 
 
vin] REALITY OF MOLECULES 119 
 
 line sections, neighbouring particles darting in quite 
 different directions as though buffeted hither and 
 thither by some hidden incalculable force. 
 
 No one at the time realised the true significance 
 of Brown's observation. But in 1863, shortly after 
 the development of the kinetic theory of heat, Wiener 
 cited the Brownian movement as a proof of the mole- 
 cular agitation in liquids, and in 1877 PP. Delsaulx 
 and Carbonelle assigned it to the same cause. 
 The latter ingeniously explained its mode of action. 
 According to the kinetic theory the pressure of a gas 
 or liquid upon any surface exposed to it is due to the 
 averaged impacts of myriads of molecules striking it. 
 A moderately small particle must be struck by such 
 a vast number of molecules per second that the mole- 
 cular pressure of the liquid averages out to be the same 
 on all sides. It is thus free to settle under the steady 
 pull of gravity, as it is observed to do. But for 
 particles beneath a certain size the case is different. 
 Their surface does not admit of enough impacts for 
 the average to hold. At any instant they may be 
 struck on one side by more molecules, or with mo- 
 mentarily greater velocities, than on the other, and 
 thus they are driven hither and thither with the 
 irregular motions observed. The smaller the par- 
 ticles, the more vivid and irregular will be their 
 motions, and this, too, accords wjith observation. 
 
 In 1888 Gouy excluded by careful experiments 
 
120 BEYOND THE ATOM [OH. 
 
 all other causes which had been suggested for the 
 Brownian movement. It could not be due to vibra- 
 tions transmitted from the ground, for it goes on 
 at night in the deeps of the country with the same 
 activity as amid the heavy traffic of a city. Nor to 
 the convection currents which make the motes dance 
 in a sunbeam, for these may be excluded without 
 affecting the motions. Moreover, the motes move in 
 streams showing the course of the currents, while the 
 most striking characteristic of the Brownian move- 
 ment is that two particles, however near, are subject 
 to motions which have no apparent relation to each 
 other. Lastly, although light is now known to exert 
 a minute pressure which may become important in 
 the case of very small particles, it cannot be con- 
 cerned here, since the light used for illumination may 
 be suddenly varied a thousandfold without effect. 
 
 The strong presumption afforded by these re- 
 searches, that in the Brownian movement the mole- 
 cules may be seen visibly at work, has been converted 
 into a certainty by Perrin's quantitative study of the 
 phenomenon since 1908. Conversely, he has been 
 able to deduce verifications of some of the most 
 abstruse mathematical results of the kinetic theory ; 
 and values for the number, size, and charge of the 
 atoms which agree very closely with those obtained 
 from the theory of radio-activity. 
 
 The labours of the founders of thermodynamics 
 
vin] REALITY OF MOLECULES 121 
 
 (Clausius, Maxwell, Gibbs, Boltzmann) lead through 
 some of the most difficult regions of modern mathe- 
 matics to certain famous theorems theorems which 
 are necessarily statistical, since among the jostling 
 crowd of molecules every variety of movement must 
 occur many millions of times every second. One is, 
 that the mean energy of the molecules of a gas or 
 liquid in equilibrium is shared equally between all 
 the different types of motion open to the molecules. 
 For instance, the mean energy of translation in any 
 direction is equal to the mean energy of rotation 
 about any axis. Another gives the law of distribu- 
 tion of velocities among the individual molecules. 
 It appears that the number of molecules which at 
 any instant have a speed differing from the mean 
 speed by a certain amount is fixed by Laplace's 
 formula in the Theory of Errors, and corresponds to 
 the number of bullets, fired at a target under similar 
 circumstances, which will be found at a given distance 
 from the bull's eye. 
 
 The equations and formulae which express these 
 propositions accurately are too complicated to be 
 understood, except by mathematicians, and a veri- 
 fication would seem at first sight out of the question. 
 But Perrin betakes himself to a practical consequence 
 of the theorems. Just as Laplace had deduced the law 
 of distribution of molecules in the atmosphere under 
 gravity, showing that for equal heights of ascent the 
 
122 BEYOND THE ATOM [CH. 
 
 number of molecules (and therefore the density and 
 barometric pressure) must fall off in geometrical 
 progression, so Perrin, on the assumption that the 
 Brownian particles behave among the molecules of 
 the liquid exactly as molecules of a larger size, 
 deduced a similar formula giving the same expo- 
 nential law of distribution. But there was this 
 difference, that, whereas Laplace's law of atmospheric 
 pressure could be verified by easy observations taken 
 over a vertical height of several miles, Perrin had to 
 carry out the whole of his experiments upon a layer 
 of liquid one-tenth of a millimetre deep ! 
 
 The first experimental difficulty was to obtain an 
 emulsion containing grains of the requisite fineness 
 and of uniform size. Two substances were employed, 
 gamboge and mastic. From these by an ingenious 
 process of ' fractional centrifugation ' emulsions were 
 prepared of uniform fineness, with grains whose 
 radius might at will be made anything from one- 
 tenth of a micron a micron is the thousandth of 
 a millimetre to 12 or 13 microns. A thin layer 
 of emulsion, about 100 microns or one-tenth of a 
 millimetre deep, was placed in a shallow glass cell 
 under a high-power microscope. With the high power 
 employed there is very little depth of focus, so that 
 only one level of the liquid could be studied at a 
 time. But this level could be varied by raising or 
 lowering the microscope by the micrometer screw, 
 
vm] REALITY OF MOLECULES 123 
 
 which allowed an adjustment to within a quarter of 
 a micron. The counting presented difficulties, for 
 not only are all the grains in rapid irregular motion 
 horizontally while they remain in the plane of vision, 
 but the slightest vertical movement upwards or 
 downwards causes them to disappear instantly, while 
 swarms of other grains, rising or falling into the plane 
 of vision, flash into view in their turn. The field was 
 therefore limited by a diaphragm pierced with a fine 
 needle, so that not more than five or six grains were 
 in view at any one instant. These the eye could take 
 in at a glance when the field was instantaneously 
 illuminated at intervals of 10 or 15 seconds, and 
 observations were thus carried on at different levels, 
 in some series as many as 13,000 grains being 
 counted. 
 
 It was found that within these narrow limits of 
 working the exponential law of distribution held 
 good, as it does for the molecules of the atmosphere. 
 How accurately, may be seen from the results of two 
 sample experiments. 
 
 With grains of radius 0*14 micron in a layer of 
 liquid 110 microns deep, the grains were counted 
 at five levels taken every 25 microns, the highest 
 level being five microns below the surface of the 
 liquid. The numbers found, reckoning downwards, 
 were proportional to 
 
 100, 116, 146, 170, 200, 
 
124 BEYOND THE ATOM [CH. 
 
 whereas the strict law would require 
 100, 119, 142, 169, 201. 
 
 With larger grains (radius 0*212 micron) so as to 
 give a more rapid diminution of density, reckoning 
 upwards at successive vertical heights of 30 microns, 
 the numbers were, as 
 
 100, 47, 22'6, 12, 
 while the law requires 
 
 100, 48, 23, ll'l. 
 
 The close agreement of these observed numbers 
 with theory, showing that so far as the law of dis- 
 tribution is concerned the Brownian grains behave 
 exactly like molecules of a larger size, is in itself 
 remarkable. Still more striking are the results when, 
 conversely, it is sought to determine the mean energy 
 of the grains, which, if the Brownian movement is 
 really the result of molecular bombardment, should 
 come out identical with the mean molecular energy 
 upon the kinetic theory, whatever the size of grain 
 employed, and by implication yield another determi- 
 nation of a fundamental physical quantity known as 
 Avogadro's constant, that is, the number of molecules 
 contained in a cubic centimetre of any gas at standard 
 temperature and pressure. This may be done. For 
 the exact equation expressing the law of distribution 
 of the grains involves the mean granular energy 
 
vni] REALITY OF MOLECULES 125 
 
 together with the radius and density of the grains, 
 so that if the radius and density can be measured, the 
 equation will give the mean energy. 
 
 Perrin measured the density by two different 
 methods with results which were concordant within 
 1 or 2 parts in a thousand. Thus, with mastic one 
 method gave a density T063, the other T064. The 
 radius was determined by two methods of direct count- 
 ing and measurement, and also by applying Stokes' 
 formula to the rate of fall of a cloud of grains through 
 the emulsion, the method employed by J. J. Thomson 
 and C. T. R. Wilson (Chap. II) for counting the 
 droplets in a charged cloud. Since the values ob- 
 tained were again highly concordant, an incidental 
 verification is afforded for a method used by many 
 investigators in recent years. 
 
 It now becomes interesting to see whether on sub- 
 stituting these values in the equation (1) the granular 
 energy comes out the same for grains of different 
 sizes and densities, and (2) the value of Avogadro's 
 constant is in the neighbourhood of 272 x 10 19 , as 
 indicated by previous researches in many different 
 departments of physics. Five series of observations 
 were made, with grains varying in mass over a range 
 of 1 to 40. In each series many thousands of grains 
 were counted or photographed. The resulting values 
 of the granular energy were in close agreement, 
 thus verifying experimentally the theorem of the 
 
126 BEYOND THE ATOM [CH. 
 
 equi-partition of energy among different masses ; and 
 the consequent values of Avogadro's constant ranged 
 from 2*3 . 10 19 to 3*5 . 10 19 with a mean of 3'1 . 10 19 the 
 most probable value from the best series being 
 3-12. 10 19 . 
 
 When it is remembered that before these ex- 
 periments it might have seemed improbable there 
 should be any change of distribution within the 
 minute height of a few microns studied, and that it 
 was just as likely that all the grains would be found 
 on the bottom of the vessel ; that in the first case 
 the value of Avogadro's constant would have come 
 out zero, and, in the second infinite ; it can hardly 
 be conceived as an accident that the value ob- 
 tained should be so near that to be expected on 
 other grounds. Thus the molecular theory of the 
 Brownian movement seems to be experimentally 
 established and with it the objective reality of the 
 molecules. 
 
 The value for Avogadro's constant finally adopted 
 by Perrin is based on still another series of obser- 
 vations carried out with the utmost precision and 
 covering 13,000 grains. It is 315 , 10 19 . From this 
 may be deduced the following numbers for com- 
 parison with the table on pp. 106, 107. 
 
 Number of molecules of any gas, at 
 standard temperature and pressure, per 
 cubic centimetre 3. 1 5 . 1 19 
 
vin] REALITY OF MOLECULES 127 
 
 Charge of an electron in electro- 
 static units 4,1 1 . 10- 10 
 
 Mass of hydrogen atom l,43.10~ a 
 
 Mass of electron (1775 times less) 0,805 . 10" 27 
 Diameter of hydrogen molecule ... 2 . 10~ 8 
 
 Diameter of electron J.10~ 12 
 
 The electron is therefore about three times as 
 small compared with the hydrogen molecule, as the 
 earth is, compared with the diameter of its orbit 
 round the sun ! 
 
 The results just given are derived from a second- 
 ary consequence of the molecular theory of the 
 Brownian movement, viz. the law of distribution 
 of the grains in a vertical column of the liquid. 
 A more direct verification is possible. Although the 
 actual speed of any one grain cannot be determined, 
 owing to the irregularities of its serrated path, 
 Einstein, applying the theory of probabilities, has 
 deduced (1) a formula for the mean displacement 
 of a grain in a given interval, (2) a proof that the 
 actual displacements observed must be distributed 
 about the mean according to Maxwell's law for the 
 distribution of the velocities about the mean velocity, 
 and (3) a formula for the energy of rotation of a 
 grain about any axis, assuming from the 'equi- 
 partition of energy' that this will be the same as 
 the mean energy of translation in any direction. All 
 these points were tested. 
 
128 BEYOND THE ATOM [OH. 
 
 (1) The actual displacements of sets of 50 to 
 200 separate grains were noted at intervals of 
 30 seconds to two minutes. From a comparison 
 of the results of six sets of observations with 
 Einstein's formula, Avogadro's constant came out 
 in close agreement with previous values, 3*19. 10 19 . 
 
 (2) The displacements were found to be dis- 
 tributed about the mean in striking agreement with 
 Maxwell's target law. 
 
 (3) Although the minute grains employed in 
 these experiments were seen to be rotating, no 
 measures could be made upon them. Naturally, 
 since Einstein's formula indicates that grains of even 
 one whole micron diameter must turn on the average 
 through 100 per second. But with much larger 
 grains, of 13 microns diameter, the mean rotation 
 comes down to 14 per minute. By means of mark- 
 ings on these larger grains the angle turned through 
 per minute could be measured, and the results of 
 200 observations gave a mean rotation of 14'5 per 
 minute ! 
 
 The Blue of the Sky. 
 
 Avogadro's constant, the number of molecules 
 in a cubic centimetre of gas at standard pressure 
 and temperature, had been estimated within some- 
 what wide limits from various thermodynamic and 
 other considerations. It could be fixed much more 
 
viii] REALITY OF MOLECULES 129 
 
 exactly by various methods (Ch. VII) based on 
 counting the atoms from radium. Thus Rutherford 
 arid Geiger obtained the value 2*72. 10 19 ; Millikan 
 2-644.10 19 . Other values were obtained from the 
 theory of radiation, 3*43. 10 19 by Lorenz from the 
 distribution of energy in the spectrum, 272. 10 19 by 
 Planck from the radiation of a black body. These 
 are in fair agreement with Perrin's value, 3*15. 10 19 , 
 above. 
 
 Still another value may be obtained, from a most 
 unexpected quarter, the blue colour of the sky. 
 Fine dust-particles or water-globules, if near the 
 wave-lengths of light in size, behave quite differently 
 to the short waves of the blue and violet and the 
 longer waves of the red. The blue waves are scat- 
 tered, i.e. reflected in all directions from the particles, 
 just as ripples are reflected from the sides of a block 
 of wood floating in calm water ; while the long red 
 waves roll through them unaffected, just as a sea- 
 wave would roll on merely tossing the block as it 
 passed. Helmholtz applied this fact to explain, for 
 instance, the colours of sunrise and sunset. An 
 observer looking down a valley towards the setting 
 sun, receives rays which have travelled a great 
 distance horizontally through the atmosphere. They 
 have consequently lost much of their blue by scatter- 
 ing, and have an excess of red. On looking across 
 them at the opposite hills he receives the rays which 
 
 c. 9 
 
130 BEYOND THE ATOM [CH. 
 
 are being scattered, and sees the blue haze of distance 
 drawn like a veil across the hills ; while from 
 overhead he receives other scattered rays giving the 
 blue of the sky. Even the red of these rays passing 
 overhead may sometimes be seen. For they are bent 
 round by refraction in their long path" through the 
 atmosphere at the rim of the earth, as they would be 
 at the edge of a spherical lens, and converge to a 
 point far behind the earth from the sun. During 
 a total lunar eclipse the moon finds herself there, 
 and this ' sifted ' light is the only light she receives, 
 since the body of the earth blocks out the direct 
 rays. And hence is she covered during totality by 
 a rosy copper glow. 
 
 Lord Rayleigh showed that the molecules of the 
 air are themselves able to scatter light in this way, 
 and from the resulting formulae he was able to 
 deduce limits for the number of molecules per cubic 
 centimetre. Quite recently L. Yessot King 1 has 
 introduced terms into Rayleigh's differential equa- 
 tions which allow for absorption and self-illumination. 
 The solution shows that the theory of molecular 
 scattering can then account for the facts of sky- 
 radiation, and, when compared with observed results, 
 yields values ranging from 2'24. 10 19 to 2'51 . 10 19 for 
 Avogadro's constant. 
 
 That values so consistent should be derived from 
 
 i Proc. Roy. Soc. Nov. 7, 1912. 
 
ix] THE NEW ATOM 131 
 
 facts so remote and apparently discrete, points to 
 the deep underlying relation between the various 
 regions of physics, which, though the links cannot 
 yet always be traced in imagination, may even now 
 be expressed in equations between the formulae for 
 Avogadro's constant. 
 
 CHAPTER IX 
 
 THE NEW ATOM 
 
 IN face of the facts we have been recounting the 
 concept of the hard, unchangeable atom needs funda- 
 mental revision. Already thirty years ago Rowland 
 of Baltimore, from a consideration of the iron spectrum 
 with its thousands of lines each indicating a definite 
 rate of vibration, was led to remark that compared 
 with an atom of iron a grand piano must be a very 
 simple structure ! And now it seems that some atoms 
 can be transmuted with the emission of helium atoms 
 and electrons. Of these, therefore, an atom must in 
 some way be built up. If, as we shall see, the mass 
 of an electron is wholly electrical, its diameter can be 
 shown to be of the order 10~ 13 cm. ; that of an atom 
 is 10~ 8 cm. The electron is thus one hundred thousand 
 times smaller than the atom, and the spaces between 
 electrons perhaps one hundred million times the 
 
 92 
 
132 BEYOND THE ATOM [CH. 
 
 diameter of an electron. This suggests inevitably an 
 arrangement like a planetary system. 
 
 Again, the velocity with which a helium atom is 
 projected from radium (7, for instance, would require 
 an electric field of more than four million volts to 
 produce it. This is inadmissible, and we are driven 
 to conclude that the atom already possessed this 
 speed before disruption. Once more the planetary 
 system is suggested, the electrons and helium atoms 
 moving about each other with orbital velocities, as 
 the planets move round the sun. 
 
 Upon these ideas J. J. Thomson has built a very 
 interesting model of an atom. It had been shown by 
 Larmor that an electric charge in motion radiates 
 energy, if its motion is changing, i.e. if it is being 
 accelerated, in proportion to the square of its accele- 
 ration. A single electron moving in a circular orbit 
 is thus a great radiator, and must soon lose its energy ; 
 but Thomson showed that the loss by radiation 
 rapidly diminishes if the number of electrons moving 
 in the circle is increased. Accordingly, as a rough 
 approximation, he worked out the case of an atom 
 consisting of a central positive charge and a number 
 of electrons arranged symmetrically round it in a 
 circle, and moving about it as the planets move round 
 the sun. A model of such a ' Saturnian ' atom has 
 interesting analogies with known chemical laws : 
 
 1. Not more than a certain number of electrons 
 
ix] THE NEW ATOM 133 
 
 can continue in stable motion in one ring. If more 
 be added, the system breaks up into two rings, an 
 inner and an outer. As the number is increased, there 
 is another rearrangement in three rings, and so on. 
 Consider now the groups corresponding to 3, 1 1, 24, 40 
 and 60 electrons. Each is derived from the pre- 
 ceding group by the addition of an extra ring. For 
 instance, the group of 40 has four rings, of 
 
 3, 8, 13, 16 electrons ; 
 that of 60 has five rings : 
 
 3, 8, 13, 16, 20. 
 
 Such atoms, having so much that is common in 
 their constitution, would be expected to show a 
 certain similarity in their properties, and might well 
 correspond to the elements of recurring character 
 which stand at the heads of the vertical columns in 
 Mendeleef 's periodic table. 
 
 2. The groups may be classified in another way. 
 We may take all those which can exist with an outer 
 ring of a definite number. For instance, there are 
 nine possible stable arrangements with an outer ring 
 of 20. The smallest number with such a ring is 59 : 
 the greatest 67. The group of 59 could not lose an 
 electron without breaking up into an entirely new 
 arrangement ; this suggests an inert element of no 
 valency. The group of 60 could only lose one electron 
 
134 BEYOND THE ATOM [CH. 
 
 without breaking up, and would be like an electro- 
 positive, monovalent element. The group of 61 is 
 more stable and could not so easily lose an electron, 
 but is capable in the end of standing the loss of two. 
 It would thus be divalent and still electropositive, 
 but less so than the group of 60. 
 
 At the other end of the series in the group of 67 
 the outer ring is very stable, but if it acquires another 
 electron, it must become the smallest and least stable 
 of the new series of groups with a larger outer ring. 
 It is more likely to lose the electron again at once, 
 and, like the group of 59, would resemble an inert 
 element of no valency. The group of 66 is the most 
 electronegative of the series, but could only retain 
 one extra electron and is therefore monovalent. The 
 group 65, though less likely to acquire electrons, 
 could retain two, and would be divalent. 
 
 It is very suggestive to compare these results 
 with the known characters of such groups of elements 
 as the following: 
 
 (1) He, Li, Be, B, C, N, 0, F, Ne\ 
 
 (2) Ne, No, Mg, Al, Si, P, 8, Cl, Arg. 
 
 In each of these groups the first and last are 
 elements of no valency ; the second and last but one 
 are monovalent but electropositive and electronega- 
 tive respectively; the third and last but two are 
 divalent, electropositive and electronegative respec- 
 tively. 
 
ix] THE NEW ATOM 135 
 
 From these considerations we may even get a hint 
 of the mechanism of chemical combination. The 
 electrons are less firmly held in the electropositive 
 members of each series. When, therefore, a substance 
 with an atom corresponding to an electronegative 
 group is mixed with one which is electropositive, the 
 negative atoms would be apt to gain an electron at 
 the expense of the positive, thus setting up electric 
 forces between the two which might hold them 
 together in permanent union. 
 
 3. Lastly, some of the arrangements are stable if 
 the orbital velocity exceeds a certain critical value, 
 but must rearrange themselves if their speed falls 
 below it ; just as a spinning top continues * sleeping ' 
 till friction reduces its speed below a certain limit, 
 when it falls over into violent irregular motion. 
 
 It is thus, according to Thomson, that the a 
 particles may be shot out from radium. Although in 
 a heavy atom built of many electrons the drain of 
 energy by radiation- may be very small yet in course 
 of time the velocities must sink to the critical value, 
 the atom falls into an entirely new arrangement with 
 diminished potential energy, and the energy released 
 may suffice to eject parts of the system with the 
 violence of an explosion. 
 
 Another cause for the disruption of an atom has 
 been suggested by Lodge. Electromagnetic theory 
 indicates, as J. J. Thomson showed in 1887, that a 
 
136 BEYOND THE ATOM [CH. 
 
 moving charge of electricity acquires a certain inertia, 
 in virtue of its motion, through the grip which the 
 Faraday lines of force travelling with it have upon the 
 ether. It thus resists change of motion as if it had 
 an electrical mass independent of any material nucleus 
 on which the charge may reside. This added mass is 
 
 2 6 2 
 nearly constant for slow speeds and equal to ] -, 
 
 o O> 
 
 where e is the charge and a the radius of the sphere 
 on which it is distributed. (It was from this formula 
 that the diameter of an electron was calculated above, 
 
 2 6 2 2 e e 
 
 since a -- = - e, and both and e are known.) 
 
 3m 3m m 
 
 But as the speed of the moving charge increases 
 beyond half the velocity of light, the electrical 
 mass grows rapidly, and theoretically should become 
 infinite when the speed of light itself is attained. Of 
 course this only means that in an ether not subject to 
 disruption the speed of light radiation is the highest 
 that can be obtained. 
 
 It has been possible to submit this theory to direct 
 experiment. Radium emits electrons with speeds 
 ranging from one-fifth up to 99'8 per cent, of the 
 
 p 
 
 velocity of light ; and by measuring the ratio for 
 
 Tfb 
 
 these Kaufmann showed in 1906 that the mass of the 
 electron was at all speeds within less than one per 
 cent, of what it should be according to theory, if the 
 
ix] THE NEW ATOM 137 
 
 mass is wholly electrical. An electron is thus a 
 disembodied charge without a material nucleus. 
 
 Now Lodge points out that the loss of energy by 
 radiation must have the same efiect on the revolving 
 electrons in an atom as motion in a resisting medium 
 would have upon the planets. The orbits would 
 slowly contract but the speed would increase as they 
 drew closer in to the central body. At last, as the speed 
 of light was approached, the mass of the electrons 
 would begin to grow rapidly, as if, while a flywheel 
 is running near the safety limit, its rim could suddenly 
 be made many times as thick. The result would be 
 explosion and the projection of parts of the system 
 with the high orbital velocities they possessed at the 
 moment. 
 
 The a Rays. 
 
 Whatever weight be attached to these particular 
 suggestions, the view that the cause of disruption of 
 an atom must be sought in the atom itself is strongly 
 supported by some interesting experiments by Bragg 
 and Kleeman on the range of the a particles in air. 
 Using a very shallow testing chamber which could be 
 set at different distances from the radium employed 
 as a source, they obtained a curve showing ionization 
 plotted against distance, with four ' knees' or bends 
 in it. The four parts of the curve correspond to the 
 
138 BEYOND THE ATOM [CH. 
 
 four sources of a particles, which are found to have 
 quite definite and different ranges in air before their 
 ionizing power is extinguished. The ionizing power 
 of each particular kind increases to a maximum as 
 the chamber is removed from the source and then 
 suddenly disappears. The ranges, which are so 
 characteristic that the source may be identified by 
 the range of the a particles it emits, are as follows : 
 for a particles from 
 
 radium 3*5 cm. radium A 4'83 cm. 
 emanation 4'23 radium C 7*06 
 
 Similar values, characteristic for each source, 
 have now been found for every transformation which 
 emits a particles, and are given in the table on p. 91. 
 
 The sharpness with which the ionizing power 
 ceases may be shown to a large audience by a beautiful 
 experiment due to Soddy. A disc rendered active by 
 exposure to radium emanation and left long enough for 
 radium C to develope is placed in the centre of a flask 
 about six inches in diameter, whose inner surface has 
 been dusted with fluorescent zinc-sulphide. No light 
 is seen, for the walls are just beyond the 7 cm. range. 
 But if the air be exhausted from the flask, at the first 
 stroke of the pump the glass shines up brilliantly 
 under the bombardment of the a particles which can 
 now reach it through the rarer air. 
 
ix] THE NEW ATOM 139 
 
 The & and y Rays. 
 
 Very interesting sidelights are thrown on this 
 view of the atom in a recent paper by Rutherford 1 
 on the origin of /3 and 7 rays. We have seen that 
 each atom of an a ray product emits on breaking 
 up a single a ray with definite range and velocity 
 characteristic for the substance, the only exception 
 being thorium (7, which emits two a rays for each 
 atom. (But see p. 93.) 
 
 By analogy it would be natural to expect that 
 one ray would be emitted for each atom. The 
 general evidence indicates that this is the case. 
 Yet the experiments of Von Baer, Hahn, Meitner 
 and Danysz have shown that the rays from the 
 same product are by no means homogeneous. When 
 a narrow pencil of rays is subjected to a magnetic 
 field, it is analysed into a ' spectrum ' consisting of 
 many bands each deflected through a different angle, 
 having therefore its own special velocity and energy. 
 Thus the rays from radium B and C together give 
 rise to as many as thirty different bands. More- 
 over, there is no obvious relation between the inten- 
 sities of the rays and the ^ rays arising from the 
 same transformation. It would seem, then, that 
 either the same type of atom may break up in many 
 totally different ways, a suggestion improbable in 
 
 1 Phil. Mag. Oct. 1912. 
 
140 BEYOND THE ATOM [OH. 
 
 itself and negatived by the fact that only one type of 
 new product results ; or, the fi rays are altered in 
 speed after their production and during their escape 
 from the atom. 
 
 Now Rutherford has found, from a study of eleven 
 out of the thirty groups of /3 rays from radium (7, that 
 the energy in each group falls short of the energy in 
 the swiftest group by a simple multiple of one or other 
 of two quite definite amounts of energy, E l and J a , or 
 by the sum of such multiples. Can we give a meaning 
 to E l and EJ. Well, Barkla, it will be remembered, 
 has shown that each element emits one or more 
 definite types of X radiation which are characteristic 
 of it ; and a comparison of his formulae with Soddy's 
 experimental results makes it probable that the 
 penetrating 7 rays from radium C are Barkla's 
 characteristic radiations from radium when excited 
 by the escape of /3 rays. Now E z turns out to be 
 very nearly the calculated energy of fi particles 
 required to excite the characteristic radiation from 
 radium (7. It is possible, though not yet proved, 
 that EI may be similarly connected with Barkla's 
 second type of radiations characteristic of radium. 
 
 Rutherford therefore conceives the atom, like 
 Thomson, as a positive nucleus (but of very small 1 
 
 1 The nucleus must be small to account for the phenomena of 
 scattering of a particles by matter. On collision with an atom the 
 a particle dives round the nucleus in a hyperbolic orbit, just as non- 
 
ix] THE NEW ATOM 141 
 
 dimensions) surrounded by an electronic system pro- 
 bably rotating in rings. Instability, for any cause, of 
 the nucleus may cause the expulsion of an a ray; 
 instability of the electrons may start a ft ray. When 
 an electron escapes from a ring near the surface of 
 the atom, it is merely a high-speed ft ray emitted 
 without production of 7 rays. Radium E and Uran- 
 ium X are cases in point. But an electron escaping 
 from an inner ring may have to pass, on its way, 
 through rings or regions where it gives rise to 7 rays, 
 with a loss of definite amounts of energy character- 
 istic for each ring. This fits well with Bragg's view 
 that ft and 7 rays are mutually convertible -forms of 
 energy, except that apparently a ft ray may give rise 
 to several 7 rays instead of the single one allowed by 
 Bragg. The same concept is applied by Sir Joseph 
 Thomson in his latest study of the positive rays in a 
 vacuum tube 1 . He finds from the photographic 
 curves that atoms may be multiply charged. The 
 hydrogen atom always carries one ionic charge. The 
 majority of the elements have twice the charge, and 
 the mercury atom has been found with as many as 
 eight ionic charges. This suggests that an atom may 
 be ionized in either of two ways : 
 
 recurring comets do in their single visit to the sun. The number of 
 a particles which will be deflected through a given angle on scattering 
 may be calculated by the theory of probabilities, and the results have 
 been verified by Geiger over a wide range of angles. 
 1 Phil. Mag. Oct. 1912. 
 
142 BEYOND THE ATOM [OH. 
 
 (1) by collision with a ft particle, when an 
 electron may be struck out from the system, or 
 
 (2) by collision with an atom, when the nucleus 
 runs away from the whole system of electrons, but 
 then regains some. 
 
 The strange fact that the ionizing power of an a 
 particle increases as its velocity falls off has been 
 verified directly by experiment, and is perhaps after 
 all only to be expected. We may form a rough idea 
 of what happens, if we regard a particle, or electron, 
 shot through a molecule as a Mauser rifle bullet of 
 small mass but high speed fired through a window. 
 It cuts a 'clean hole and removes but a minute piece 
 of glass. The a particle is by comparison a large and 
 lumbering object which shatters the molecule as a 
 brick hurled through the window would shatter the 
 glass. And just as the brick would be more effective 
 the lower its velocity until its speed is just too small 
 to break the glass, so the a particle, always a far more 
 efficient ionizer than the small and speedier electron, 
 will increase in efficiency till suddenly it loses its 
 power altogether \ 
 
 i C. T. E. Wilson (Proc. Roy. Soc. A. 87, 1912) has actually 
 photographed the trails of both a and /S particles, shown by the 
 droplets condensed on the ions produced on their passage. The 
 first photograph (see Plate) shows the trails of two a particles. They 
 are continuous, owing to the large number of ions produced ; and 
 they are deflected , through 10 and 43 respectively by ' single 
 scattering 9 on collision with an atom. In the second photograph 
 
I 
 
ix] THE NEW ATOM 143 
 
 That the range should be a fixed property of the 
 source of a particles shows that the initial velocity of 
 discharge is a fixed character.of the atoms disintegrat- 
 ing, not lost even in the act of bursting. 
 
 Some consequent experiments by Rutherford 
 showed that when the ionizing power was lost, the 
 photographic and fluorescent properties went with it 
 and yet the speed of the a particles had only been 
 reduced by about 40 per cent. They still retained 
 about one-twentieth of the speed of light. There 
 seems to be, therefore, a critical velocity below which 
 the projected a particles lose the power of producing 
 all the effects by which alone they can be detected 1 . 
 J. J. Thomson suggests that when the speed sinks to 
 this critical value the positive a particles can no 
 longer prevent an electron from tacking on to them. 
 They thus become neutral, and henceforth differ in no 
 respect from an ordinary atom of helium. 
 
 It is impossible to resist the thought that if this 
 be so, all the known kinds of ordinary matter may be 
 undergoing slow transformations, but that only those 
 which, like the radio-active elements, give rise to 
 
 the trails of the /3 particles are beaded, for only a few ions are produced, 
 strung out along the trails. The writer has to thank Mr Wilson for 
 kindly allowing the use of these photographs. 
 
 1 But Geiger, using the method of Scintillations, has since detected 
 a particles with only *26 of the velocity of a particles from radium C. 
 Though the speed diminishes rapidly towards the end of the path 
 there is no certain evidence of a critical velocity. 
 
144 BEYOND THE ATOM [OH. 
 
 the projection of electrons and a particles with more 
 than a certain velocity are open to our detection. 
 This would, indeed, point to an evolution of inanimate 
 matter analogous to that evolution of types which 
 rules in the realm of organized life. 
 
 But here is met again the question which led 
 Herschel in the seventies to class the atoms as 
 * manufactured ' articles. Why are there only some 
 hundred distinct types? And how is it that the 
 myriad members of each type resemble each other 
 with such exactitude that an atom of hydrogen, for 
 example, on the confines of the universe can be 
 recognized and identified with its kindred on earth 
 to-day by the momentary quiver it imparted to the 
 ether thousands of years ago? From the point of 
 view just indicated the answer is this. The known 
 elements are but the more stable stages in the uni- 
 versal flux, with a mean life so long that they ' abide 
 our question' by the gross methods of ordinary 
 chemistry. They are in fact, by a further analogy 
 with the evolution of the animate world, the survival 
 of the fittest among the infinitely varying types of 
 matter, the fittest to survive. 
 
 But it is time to draw a sharp line between the 
 speculations sketched in this chapter and the solid 
 experimental facts described in those which precede 
 it. We have reached the verge where knowledge 
 is in the making. It is permissible to take a look 
 
ix] THE NEW ATOM 145 
 
 forward with the pioneers into the unknown ; but 
 safer to get back as soon as may be to established 
 fact. As Leonardo, that wonderful forerunner, said 
 four hundred years ago : 
 
 ' Nature never deceives. Only man's judgment 
 deceives, in promising things not supported by ex- 
 periment/ 
 
 What the next step will be, whether all mass 
 will be proved to be electrical like that of the 
 electron, or whether it be some great simplification 
 by means of a universal concept such as the ether 
 that makes the interest of the outlook. 
 
 Meanwhile, it is certain that in some cases trans- 
 mutation, the dream of the alchemist, goes slowly but 
 inexorably on, accompanied by the unlocking of vast 
 stores of atomic energy. But by a strange irony the 
 interest has shifted. The alchemist sought to turn 
 the baser to the nobler metals by the sacrifice of 
 energy in the shape of heat, but was impotent to 
 effect the change. His successor of the twentieth 
 century knows that if the transmutation could be 
 effected the metals would be but as dross to the stores 
 of energy set free, but is equally impotent to delay or 
 hasten the change. 
 
 The new knowledge gained in the last decade is a 
 true development, not a destructive revolution. Con- 
 servation of Energy stands where it did, though un- 
 suspected and enormous stores of atomic energy have 
 
 c. 10 
 
146 BEYOND THE ATOM [OH. ix 
 
 been brought into view. And the Atomic Theory is 
 still the guide of the chemist, though for the purposes 
 of modern physics he has been compelled to look 
 1 beyond the atom,' and in the words of Blake 
 
 ' To see a world in a grain of sand 
 
 And a heaven in a flower. 
 To grasp infinity in the palm of the hand, 
 And eternity in an hour.' 
 
BIBLIOGRAPHY 
 
 Books. 
 
 E. Rutherford. Radio- Activity. Camb. Univ. Press. 2nd ed. 
 
 1905. 
 Radioactive Substances and their Radiations. Camb. Univ. 
 
 Press, 1913. 
 
 Radio- Active Transformations. Lond. 1906. 
 M. Curie. Traite de Radioactivite. Paris, 1910. 
 R. J. Strutt. The Becquerel Rays and the Properties of 
 Radium. Lond. 1904. 
 
 F. Soddy. The Interpretation of Radium. Lond. 1909. 
 
 Scientific Papers. 
 J. J. Thomson. Kathode Rays. Philosophical Magazine, 
 
 Oct. 1897. 
 The Charge of Electricity carried by the Ions produced by 
 
 Rontgen Rays. Phil. Mag. 1898, XLVI. p. 528. 
 The Structure of the Atom. Phil. Mag. Mar. 1904. 
 Rays of Positive Electricity. Phil. Mag. Feb. 1911. 
 A New Method of Chemical Analysis. Chemical News, 
 
 June 9, 1911. 
 
 W. H. Bragg. The Consequences of the Corpuscular Hypothesis 
 of the y and X Rays, and the Range of ft Rays. Phil. Mag. 
 Sept. 1910. 
 C. G. Barkla. Typical Cases of lonization by X Rays. Phil. Mag. 
 
 Aug. 1910. 
 
 Dr R. W. Gray and Sir W. Ramsay. The Density of Radium 
 Emanation and the Disintegration Theory. Chem. News, 
 Feb. 24 and Mar. 3, 1911. 
 
148 BIBLIOGRAPHY 
 
 E. Rutherford and F. Soddy. Radio-active Change. Phil Mag. 
 1903. 
 
 E. Rutherford and H. Geiger. 1. An Electrical Method of 
 Counting the number of a particles from Radio-active Sub- 
 stances. 2. The Charge and Nature of the a Particle. Proc. 
 Roy. Soc. A. 81. 1908. 
 
 E. Rutherford. The Origin of /3 and y Rays from Radio-active 
 Substances. Phil. Mag. Oct. 1912. 
 
 J. Perrin. Mouvement Brownien et Realite Moleculaire. 
 Annales de Chimie et de Physique. 
 
INDEX 
 
 a particles, 
 
 helium suggested 73 
 
 are helium atoms 104 
 
 experimental proof 105 
 
 number per second 98, 102 
 
 ionic charge from 103 
 
 scattering 140 
 
 critical velocity 143 
 a, /3, 7 rays, penetrating power 
 
 47 
 o rays, 
 
 magnetic deflection 48 
 
 nature 50 
 
 range 91, 138 
 Actinium, discovery 46 
 Active deposits, 
 
 discovery 63 
 
 properties 65 
 
 curves 66 
 
 analysis of curves 83-85 
 
 rapid change 80 
 
 a ray curves 81 
 
 ft ray curve 82 
 
 slow change 86 
 Age of earth 109 
 
 rocks 111, 112 
 
 sun 112 
 
 Analogy of a, , y rays to vacuum- 
 tube 51 
 Atom, 
 
 Maxwell 13 
 
 model, Thomson 132 
 
 Atom, cont. 
 
 Rutherford 140 
 
 electro-positive and electro- 
 negative 133-4 
 
 cause of disruption 135 
 
 vortex theory 11 
 Atom of electricity 30 
 Atomic charge 25 
 
 Thomson 28 
 
 Millikan 29 
 
 Ehrenhaft 30 
 
 de Broglie 30 
 Atomic constants, 
 
 Maxwell 10 
 
 from radio-activity 106-7 
 
 from Brownian movement 126 
 Atomic structure of electricity 25 
 Atomic Theory 6 
 Aurora 115 
 Avogadro's constant 129 
 
 from Brownian movement 126 
 
 from blue of sky 130 
 Avogadro's Law 11 
 
 ft rays, e\m 50 
 
 nature of 50, 139 
 
 velocity 50, 75 
 Blue of distance 129 
 
 of sky 129 
 
 Balance, Steele and Grant 95 
 Brownian movement 118 
 
 atomic constants from 126 
 
150 
 
 INDEX 
 
 Brownian particles, 
 distribution 123-4 
 rotation 128 
 
 Canal rays 31 
 
 nature of 32 
 Chalcolite 44 
 Characteristic radiations (Barkla) 
 
 140 
 Chemical combination, nature of 
 
 135 
 Cloud-formation 26 
 
 rate of fall 26 
 Comets' tails 114 
 Counting the atoms, 
 
 electrical method 100 
 
 scintillations 102 
 
 on diamond plate 104 
 Crookes' tubes 16 
 
 Disintegration theory 67 
 Double disintegration 93 
 
 Electrons 31 
 
 size of 127 
 Elements 144 
 Emanating power 62 
 Emanations, 
 
 discovery 59 
 
 condensation 61 
 
 heat from 109 
 
 nature of 61 
 
 properties 60 
 
 rate of production 62 
 
 volume produced 63, 97, 105, 
 106 
 
 Kathode rays, e/m 20, 22 
 
 Zeeman effect 24 
 Equilibrium, radio-active 70 
 Etna, lava from 110 
 
 Fergusonite 111 
 
 Gas Laws 9 
 7 rays 47 
 
 accompany /3 rays 75 
 
 nature 50 
 
 Half- value, fall to 58 
 Heat, emission by radium 108 
 Helium 111 
 rate of production 106 
 
 Imponderables 4 
 Ionic charge 30, 103, 104 
 Ionium 90 
 lonization 36, 142 
 Ions 22 
 
 properties 27 
 Iron spectrum 131 
 Isolation of radium JB and C 
 87 
 
 Kathode rays, 
 charge on (ferrin) 18 
 in air (Lenard) 18 
 nature (Thomson) 19, 31 
 
 Kinetic theory 6 
 
 Lead 94 
 
 Lunar eclipse 130 
 
 Magnetic storms 116 
 Mass, electrical 136 
 Maxwell on atoms 13 
 Medicinal springs 112 
 Mendeleefs table 133 
 Molecular weight of emanation 
 
 95 
 
 Molecules, reality of 117 
 Moon, colour in eclipse 130 
 
 Nascent state 33 
 
 New method of chemical analysis 
 
INDEX 
 
 151 
 
 Ordinary matter, possibly radio- 
 active 143 
 
 Penetrating power of a, /Sprays 47 
 Pitchblende 44, 45 
 Planetary system 132 
 Polonium 45, 93 
 Positive rays 141 
 Prout's hypothesis 31 
 
 Kadiant matter 18 
 Eadio-activity, 
 
 discovery 43 
 
 uranium 44 
 
 thorium 44 
 Eadio-active lead 46 
 
 is radium D 94 
 Eadio-tellurium 46 
 Kadium, 
 
 discovery 45 
 
 not a compound 76 
 
 spectrum 46 
 
 growth of 89 
 
 final product lead 94 
 Eadium G, 
 
 swift a, /3 rays 87 
 
 double break-up 88 
 Eadium clock 51 
 Eayless change 84, 87 
 Eays, three types, 
 
 in vacuum-tube 42 
 
 from radium 51, 52 
 
 of positive electricity 33 
 Eeality of molecules 117 
 Eecoil of atoms 107 
 Eoentgen rays 34 
 
 pulse theory (Stokes) 38 
 
 neutral pairs (Bragg) 41 
 
 characteristic spectrum(Barkla) 
 
 Simplon tunnel 110 
 Spinthariscope 99 
 Springs, medicinal 112 
 Sun, age of 112 
 Sunset colours 129 
 
 Tables, 
 
 atomic constants 10 
 
 from Brownian movement 
 126-7 
 
 from radio-activity 106-7 
 
 radium family 91 
 
 thorium family 92 
 
 actinium family 92 
 Thermodynamics 127-8 
 Thorianite 111 
 Thorium X, 
 
 discovery 53 
 
 curves 57 
 
 Trails of a, j8 particles 142 
 Transmutation 77, 145 
 
 Ultra-violet light 25 
 Uranium X, 
 
 discovery 53 
 
 curves 55 
 
 Vacuum-tube 15 
 Volume of emanation 77, 
 
 helium 106 
 Vortex theory of atom 11 
 
 106 
 
 X-rays 34 
 properties 35 
 pulse theory (Stokes) 38 
 neutral pairs (Bragg) 41 
 characteristic spectrum(Barkla) 
 41 
 
 Zeeman effect 24 
 
 CAMBRIDGE : PRINTED BY JOHN CLAY, M.A. AT THE UNIVERSITY PRI 
 
THIS BOOK IS DUE > I THE LAST DATE 
 STAMPED BELOW 
 
 AN INITIAL FINE OF 25 CENTS 
 
 WILL. BE ASSESSED FOR FAILURE TO RETURN 
 THIS BOOK ON THE DATE DUE. THE PENALTY 
 WILL INCREASE TO SO CENTS ON THE FOURTH 
 DAY AND TO $1.OO ON THE SEVENTH DAY 
 OVERDUE. 
 
 m FE3 281937 
 
 gsH 
 
 StP&8t963 
 
 
 il IVlA^ I] 1QQ7 
 
 
 gOApr'SSLA 
 
 
 REC'D LD 
 
 
 nr.T9 '63-12M 
 
 REC'P |J-> 
 
 DEC 1 2 |967 
 RECEIVED 
 
 , 
 
 mi 
 
 
 -4 ^M 
 
 N(W28'6/-IOPW 
 
 . i s\i-r\ *-* 
 
 LOAN DE^T. 
 
 ^|g^? 
 
 -^ 
 
 ml 
 
 
 ^ 
 
 ^nd'^SS 
 
 <& 
 
 
 1 
 
 -90f** 
 
U. C. BERKELEY LIBRARIES 
 
 CDbl3t,7b75 
 
 UNIVERSITY OF CALIFORNIA LIBRARY