YALE UNIVERSITY 
 MRS. HEPSA ELY SILLIMAN MEMORIAL LECTURES 
 
 RADIOACTIVE TRANSFORMATIONS 
 
SILLIMAN MEMORIAL LECTURES 
 
 PUBLISHED BY CHARLES SCRIBNER'S SONS 
 
 ELECTRICITY AND MATTER 
 
 By PROF. J. J. THOMSON. Net $1.25 
 
 THE INTEGRATIVE ACTION OF THE 
 NERVOUS SYSTEM 
 
 By PROF. C. S. SHERRINGTON. Net $3.50 
 
 RADIOACTIVE TRANSFORMATIONS 
 
 By PROF. E. RUTHERFORD. Net $3.50 
 
RADIOACTIVE 
 TRANSFORMATIONS 
 
 BY 
 
 E. RUTHERFORD, D.Sc., LL.D., F.R.S. 
 
 Mac dona Id Professor of Physics, Me Gil I University, 
 Montreal 
 
 WITH DIAGRAMS 
 
 OF THE 
 
 UNIVERSITY 
 
 OF 
 
 CHARLES SCRIBNER'S SONS 
 NEW YORK 1906 
 

 Copyright, 1906, 
 BY YALE UNIVERSITY. 
 
 Published September, 1906 
 
 THE UNIVERSITY PRESS, CAMBRIDGE, U.S.A. 
 
THE SILLIMAN FOUNDATION 
 
 IN the year 1883 a legacy of eighty thousand dollars 
 was left to the president and Fellows of Yale College in 
 the city of New Haven, to be held in trust, as a gift 
 from her children, in memory of their beloved and hon- 
 ored mother Mrs. Hepsa Ely Silliman. 
 
 On this foundation Yale College was requested and 
 directed to establish an annual course of lectures de- 
 signed to illustrate the presence and providence, the 
 wisdom and goodness of God, as manifested in the natural 
 and moral world. These were to be designated as the 
 Mrs. Hepsa Ely Silliman Memorial Lectures. It was 
 the belief of the testator that any orderly presentation 
 of the facts of nature or .history contributed to the end 
 of this foundation more effectively than any attempt to 
 emphasize the elements of doctrine or of creed ; and he 
 therefore provided that lectures on dogmatic or polemical 
 theology should be excluded from the scope of this foun- 
 dation, and that the subjects should be selected rather 
 from the domains of natural science and history, giving 
 special prominence to astronomy, chemistry, geology, and 
 anatomy. 
 
 It was further directed that each annual course should 
 be made the basis of a volume to form part of a series 
 constituting a memorial to Mrs. Silliman. The memo- 
 rial fund came into the possession of the Corporation 
 of Yale University in the year 1902; and the present 
 volume constitutes the third of the series of memorial 
 lectures. 
 
 i 
 
PREFACE 
 
 THE present work contains the subject matter of eleven 
 lectures delivered under the Silliman Foundation at 
 Yale University, March, 1905. 
 
 I chose as the subject of my lectures the most recent 
 and at the same time the most interesting development 
 of Radioactivity, namely the transformations which are 
 continuously taking place in radioactive matter. While 
 dealing fully with this aspect of the subject, it was neces- 
 sary for clearness to give some account of radioactive 
 phenomena in general, although with much less com- 
 pleteness than in my previous book on Radioactivity. 
 
 In arranging the chapters of the present volume, the 
 order in which the subject was dealt with in the lectures 
 has been closely followed, but as our knowledge of the 
 subject is increasing so rapidly, I have thought it desirable 
 to incorporate the results of the many important inves- 
 tigations which have been made since the lectures were 
 delivered. This is .especially the case in the chapter 
 dealing with the a rays, to which much attention has 
 been devoted in the past year on account of the im- 
 portant part they play in radioactive transformations. 
 
 I am much indebted to my colleagues Professor Hark- 
 ness and Professor Brown for the great care and trouble 
 they have taken in the correction of the proofs and for 
 many useful suggestions. 
 
 E. RUTHERFORD. 
 
 McGiLL UNIVERSITY, 
 
 MONTREAL, June 4, 1906. 
 
CONTENTS 
 
 CHAPTER PAGE 
 
 I. HISTORICAL INTRODUCTION i 
 
 II. RADIOACTIVE CHANGES IN THORIUM 37 
 
 III. THE RADIUM EMANATION 70 
 
 IV. TRANSFORMATION OF THE ACTIVE DEPOSIT OF RADIUM . 95 
 V. ACTIVE DEPOSIT OF RADIUM OF SLOW TRANSFORMATION . 122 
 
 VI. ORIGIN AND LIFE OF RADIUM 148 
 
 VII. TRANSFORMATION PRODUCTS OF URANIUM AND ACTINIUM, 
 
 AND THE CONNECTION BETWEEN THE RADIOELEMENTS 162 
 VIII. THE PRODUCTION OF HELIUM FROM RADIUM AND THE 
 
 TRANSFORMATION OF MATTER 179 
 
 IX. RADIOACTIVITY OF THE EARTH AND ATMOSPHERE . . . 196 
 
 X. PROPERTIES OF THE RAYS . 219 
 
 XL PHYSICAL VIEW OF RADIOACTIVE PROCESSES 256 
 
 INDEX ...... 277 
 
RADIOACTIVE 
 TRANSFORMATIONS 
 
 CHAPTER I 
 HISTORICAL INTRODUCTION 
 
 THE last decade has been a very fruitful period in physical 
 science, and discoveries of the most striking interest and 
 importance have followed one another in rapid succession. 
 Although the additions to our knowledge have come from 
 investigations in very different fields, yet a close examination 
 shows that they are all intimately related, and each discovery 
 has supplied the necessary stimulus and suggestion to serve as 
 a starting point for the next advance. 
 
 The march of discovery has been so rapid that it has been 
 difficult even for those directly engaged in the investigations to 
 grasp at once the full significance of the facts that have been 
 brought to light. Especially has this been the case in the field 
 of radioactivity, where the phenomena observed have been so 
 complicated and the laws controlling them so unusual that it 
 has been necessary to introduce conceptions of a novel character 
 for their explanation. 
 
 The starting point of this epoch in physical science was the 
 discovery by Rontgen of the X-rays in 1895 and the experiments 
 of Lenard on the cathode rays. The extraordinary properties 
 of the X-rays at once focussed the attention of the scientific 
 world, and led to a series of investigations whose object was 
 not only to examine the properties of the rays themselves, but 
 to disclose their real nature and origin. 
 
 The latter problem led to a much closer investigation of the 
 nature of the cathode rays produced in a vacuum tube, for 
 these rays were seen to be in some way intimately connected 
 with the emission of X-rays. J. J. Thomson in 1897 finally 
 
 l 
 
2 RADIOACTIVE TRANSFORMATIONS 
 
 succeeded in proving definitely that the cathode rays consisted 
 of a stream of particles moving with great velocities and carry- 
 ing negative charges of electricity. These particles had an 
 apparent mass only about T ^<j^ that of the hydrogen atom, and 
 were therefore the smallest bodies known to science. These 
 "corpuscles," or " electrons, " as they have been termed, are 
 apparently a constituent of all matter, and are believed to be 
 the ultimate parts of which the atom is composed. 
 
 This electronic hypothesis has been extremely fertile, and has 
 already greatly changed or rather extended previous concep- 
 tions of the constitution of matter. It has opened up wide 
 fields of investigation in many departments of physical science, 
 and has provided science with a microscope, so to speak, with 
 which to examine the structure of the atom of the chemist. 
 J. J. Thomson has examined mathematically the stability of 
 atoms composed of a number of whirling electrons, and has 
 shown that these model atoms imitate in a remarkable way some 
 of the more fundamental properties of the chemical atom. 
 
 The proof of the corpuscular character of the cathode rays at 
 once indicated the probable explanation of the origin and nature 
 of the X-rays. Stokes, J. J. Thomson, and Weichert independ- 
 ently suggested that the cathode rays were the parents of the 
 X-rays. The sudden stoppage of the electrons in the cathode 
 stream causes an intense electromagnetic disturbance which 
 travels out from the point of impact with the velocity of light. 
 On this point of view, the X-rays consist of a number of dis- 
 connected pulses, following one another in rapid succession 
 but without any definite order. They are akin in some respects 
 to very short waves of ultra-violet light, but differ from them in 
 the lack of periodicity of the pulses. The penetrating power of 
 the rays, and the absence of any direct reflection, refraction, 
 or polarization, were consequences of this theory if the breadth 
 of the pulse was short compared with the diameter of the atom. 
 
 An admirable and simple account of the nature and properties 
 of such pulses has been given by J. J. Thomson 3 in the Silli- 
 man lectures of 1903. 
 
 1 J. J. Thomson: Electricity and Matter (Scribner, New York, 1904). 
 
HISTORICAL INTRODUCTION 3 
 
 In the meantime, another remarkable property of the X-rays 
 had been closely examined. The passage of the X-rays through 
 a gas imparts to it a new power of rapidly discharging an electri- 
 fied body. This was satisfactorily explained on the hypothesis 
 that the rays produced a number of positively and negatively 
 charged carriers or ions in the electrically neutral gas. 1 The 
 development of this subject proceeded along two distinct lines, 
 one electrical and the other optical. C. T. R. Wilson 2 found 
 that under certain conditions the ions produced in the gas by 
 X-rays become nuclei for the condensation of water upon them. 
 Each ion thus becomes the centre of a visible charged globule 
 of water which moves in an electric field. Experiments of this 
 character verified in a remarkable way the fundamental correct- 
 ness of the ionization theory, and clearly brought out the dis- 
 continuous or atomic structure of the carriers of the electric 
 charges. 
 
 As a result of researches on diffusion of the ions in erases, 
 
 o 
 
 Townsend 3 deduced the important fact that the charge carried 
 by a gaseous ion was the same in all cases, and equal to the 
 charge carried by the hydrogen atom in the electrolysis of 
 water. By a combination of the electrical and optical methods, 
 J. J. Thomson 4 found the actual value of the charge carried by 
 an ion. 
 
 The determination of this important physical unit at once 
 allows us to count the number of ions present in any volume of 
 air acted upon by an ionizing agent. In addition to this, it pro- 
 vides the most accurate deduction yet made of the number of 
 molecules present in unit volume of any gas at standard pres- 
 sure and temperature. This number, which is based purely on 
 experimental data, will be seen to be of the greatest value in 
 calculating the magnitudes of various quantities in the subject 
 of radioactivity. 
 
 The ionization theory of gases was successfully applied to 
 
 1 J. J. Thomson and E. Rutherford : Phil. Mag., Nov., 1896. 
 
 2 C. T. R. Wilson: Phil. Trans., p. 265, 1897 ; p. 403, 1899; p. 289, 1900. 
 
 3 Townsend: Phil. Trans. A, p. 129, ^SW. ?^00 ItAtffy) 
 
 4 J. J. Thomson: Phil. Mag., Dec., 1898; March, 1903. 
 
4 RADIOACTIVE TRANSFORMATIONS 
 
 account for the conductivity of flames and heated vapors and to 
 unravel the complicated phenomena observed in the discharge 
 of electricity through a vacuum tube. This fascinating and far- 
 reaching field of physical inquiry owes its inception and much 
 of its development to Prof. J. J. Thomson and his students 
 at the Cavendish Laboratory, Cambridge. 
 
 On the theoretical side the possibilities of an ionic or elec- 
 tronic theory of matter had been recognized long before the 
 experimental evidence was forthcoming. The most notable ex- 
 ponents of this school were Lorentz and Larmor, who developed 
 their theories to account, among other things, for the mechan- 
 ism of radiation. The discovery by Zeeman of the action of a 
 magnetic field in displacing the spectral lines afforded a strong 
 confirmation of the general theory, for the experimental results 
 observed were in large part predicted by the theory of Lorentz. 
 In addition it was deduced that the ion, whose movements gave 
 rise to the radiation, had a mass of about the same small value 
 as the corpuscle of J. J. Thomson observed in a vacuum tube. 
 Results of this character at once extended the range of ionic 
 theories to matter in general, and though much still remains 
 to be done, the electronic theory has already proved of great 
 value in elucidating the connection between some of the most 
 recondite physical phenomena. 
 
 The movement set on foot by the discovery of Rontgen had 
 even more important consequences in another very unexpected 
 direction. Immediately after the discovery of the X-rays, it 
 was thought that the emission of these rays was in some way 
 connected with the phosphorescence set up by the cathode rays 
 on the walls of a vacuum tube. 
 
 It occurred to several scientists that natural bodies which 
 phosphoresced under the influence of light might possess 
 the property of emitting a penetrating type of radiation 
 similar to X-rays. We now know that this speculation 
 had no secure basis in fact, but it provided the stimulus 
 for the investigation of the properties of substances in this 
 special direction and soon led to a discovery of far-reaching 
 importance. 
 
HISTORICAL INTRODUCTION 5 
 
 M. Henri Becquerel, 1 a most distinguished French physicist, 
 in pursuance of this idea exposed amongst other substances a 
 phosphorescent uranium compound the double sulphate of 
 uranium and potassium to a photographic plate enveloped in 
 black paper. A darkening of the plate was observed, showing 
 that this substance emitted rays capable of passing through 
 matter opaque to ordinary light. It was soon found, however, 
 that this property was not in any way the result of phosphores- 
 cence, for it was exhibited by all the compounds of uranium 
 and the metal itself, even if these had been kept for a long time 
 in a dark room. 
 
 The radiations emitted from uranium were found to be similar 
 to X-rays in their penetrating powers. It was at first thought 
 that they differed from X-rays in showing some evidence of 
 reflection, refraction, and polarization, but this was found later 
 to be incorrect. 
 
 Becquerel observed that the uranium rays, in addition to their 
 photographic action, possessed, like X-rays, the important prop- 
 erty of discharging an electrified body. This was later examined 
 in detail by the writer, 2 who found that this discharging action 
 could be explained on the assumption that the gas was ionized 
 by the passage of the radiations through it. The ions were 
 found to be identical with those produced by X-rays, and the 
 ionization theory could consequently be directly applied to ex- 
 plain the various discharge phenomena produced by the rays 
 from uranium. At the same time, it was clearly brought out 
 that the rays from uranium consisted of two distinct kinds, 
 called the a and & rays. The former were very easily absorbed 
 in air and in thin sheets of foil, while the latter were of a far 
 more penetrating type. 
 
 The intensity of the radiations emitted by uranium, whether 
 examined by the photographic or electrical method, remains con- 
 stant, or at any rate changes extremely slowly, for no appreciable 
 alteration has been observed over a period of several years. 
 The photographic and electrical effects exhibited by uranium 
 
 1 Becquerel: Comptes rendus, cxxii, pp. 420, 501, 559, 689, 762, 1086 (1896). 
 
 2 Rutherford: Phil. Mag., Jan., 1899. 
 
6 RADIOACTIVE TRANSFORMATIONS 
 
 are very feeble compared with those produced by the X-rays 
 from the ordinary focus tube. An exposure to uranium salts 
 of at least one day is required to produce any marked action on 
 the plate. 
 
 The term "radioactivity" is now generally understood to 
 signify the property shown by the class of substances, of which 
 uranium, thorium, and radium are the best-known examples, 
 of spontaneously emitting special types of radiations capable of 
 acting on a photographic plate and of discharging an electrified 
 body. The term " activity " is used to denote the intensity of 
 the electrical or other effect of the rays from a substance com- 
 pared with that shown by some standard substance. Uranium 
 is usually chosen as this standard substance, in consequence of 
 the constancy of its radiations, and the activity exhibited by 
 other bodies is usually expressed in terms of the ratio of the elec- 
 trical effect produced by the active matter under consideration 
 compared with that of an equal weight either of uranium metal 
 itself or of uranium oxide spread over an equal radiating area. 
 For example, when the activity of radium is said to be about 
 two millions, it is meant that the electrical effect due to it is 
 two million times as great as the corresponding effect produced 
 by an equal weight of uranium spread over an equal area. 
 
 Although the property possessed by uranium of spontaneously 
 emitting energy in special forms without any apparent change 
 in the matter itself could riot fail to be regarded as a most re- 
 markable phenomenon, yet the rate of emission of energy, judged 
 by ordinary standards, is so feeble that it did not attract that 
 active scientific attention which was afterwards excited by the 
 discovery of radium ; for this substance exhibited the properties 
 of uranium to such a remarkable degree that it impressed both 
 the lay and- the scientific mind. 
 
 Shortly after the discovery of Becquerel, Mme. Curie 1 made a / 
 systematic examination of different substances for radioactivity, 
 and found that the element thorium possessed a similar property 
 to uranium and almost to the same degree. This fact was also 
 
 1 Mme. Curie: Comptes rendus, cxxvi, p. 1101 (1898). 
 
HISTORICAL INTRODUCTION 7 
 
 Independently observed by Schmidt. 1 An examination was 
 then made of the natural minerals which contain thorium and 
 uranium, and here an unexpected result was observed. Some 
 of these minerals were found to be several times more radioac- 
 tive than pure uranium or thorium, and in all cases the uranium 
 minerals showed four to five times the activity to be expected 
 from the percentage of uranium present. Mme. Curie found 
 that the radioactivity of uranium was an atomic property, i. e., 
 the activity observed depended only on the amount of the ele- 
 ment uranium present, and was not affected by its combination 
 with other substances. This being so, the large activity of the 
 uranium minerals could only be accounted for by supposing that 
 another unknown substance was present, which was far more 
 active than uranium itself. 
 
 Relying on this hypothesis, Mme. Curie boldly proceeded to 
 see if it were possible to separate chemically this unknown active 
 substance from uranium minerals. By the courtesy of the Aus- 
 trian Government, she was presented with a ton of uranium 
 residues from the State Manufactory at Joachimsthal, Bohemia. 
 In this locality there are extensive deposits of uraninite, com- 
 monly called pitchblende, which are mined for the uranium they 
 contain. This pitchblende consists mainly of uranium, but also 
 contains small quantities of a number of rare elements. 
 
 As a guide to the separation of the active substance, Mme. 
 Curie employed a suitable electroscope to measure the ioniza- 
 tion produced by the active body. After any chemical separation, 
 the activities of the precipitate and of the filtrate evaporated to 
 dryness were separately examined, and in this way it was possi- 
 ble to ascertain whether the active substance had been mainly 
 precipitated or left behind in the filtrate. 
 
 The electric method was thus used as a rapid means of qualita- 
 tive and quantitative analysis. Proceeding in this way, Mme. 
 Curie found that not one but two very active substances were 
 present in the uranium residues. The former of these, which 
 was separated with bismuth, was called polonium, 2 in honor of 
 
 1 Schmidt: Annal. d. Phys., Ixv, p. 141 (1898). 
 
 2 Mme. Curie: Comptes reudus, oxxvii, p. 175 (1898). 
 
8 RADIOACTIVE TRANSFORMATIONS 
 
 the country of her birth, and the latter, which was separated with 
 barium, was called radium. 1 This latter name was a happy in- 
 spiration, for the activity of this substance in a pure state is at 
 least two million times that of uranium. Mme. Curie then 
 proceeded with the laborious work of separating the radium 
 from the barium, and was finally successful in obtaining a small 
 quantity of probably pure radium chloride. The atomic weight 
 was found to be 225. The spectrum, first examined by De- 
 margay, was found to consist of a number of bright lines, and 
 was analogous in many respects to the spectra of the alkaline 
 earths. 
 
 In chemical properties radium is closely allied to barium, 
 but can be completely separated from it by taking account of 
 the differences in the solubility of the chlorides and bromides. 
 On account of the small quantities of radium compounds avail- 
 able and of their great cost, no one has yet endeavored to obtain 
 radium in a metallic state. Marckwald, 2 however, electrolyzed 
 a radium solution with a mercury cathode, and concluded that 
 the metal forms an amalgam with the mercury in the same way 
 as barium. The small trace of the metal so obtained exhibited 
 the characteristic radiating properties of the radium compounds. 
 
 There cannot be the slightest doubt that when radium is 
 obtained in the metallic state it will be radioactive; for the 
 property of radioactivity is atomic and not molecular. In 
 addition, uranium and thorium as metals exhibit the activity 
 to be expected from an examination of the activities of their 
 compounds. 
 
 Radium exists in very small quantity in radioactive minerals. 
 It will be seen later that the amount of radium in different 
 minerals is always proportional to their content of uranium. 
 The amount of radium per ton of uranium is about .35 gram, 
 or less than one part in a million of the mineral. From a 
 ton of Joachimsthal uraninite, which contains about 50 per 
 cent of uranium, the theoretical yield of radium should be about 
 .17 gram. 
 
 1 M. and Mme. Curie and G. Bemont: Comptes rendus, cxxvii, p. 12T5 (1898). 
 
 2 Marckwald : Ber. d. d. chem. Ges., No. 1, p. 88, 1904. 
 
HISTORICAL INTRODUCTION 9 
 
 In order to separate the radium from the barium mixed with 
 it, Mme. Curie employed the method of fractional crystalliza- 
 tion of the chlorides. Giesel 1 found that by use of the bromide 
 instead of the chloride the separation of radium from barium 
 was much facilitated. He states that six crystallizations suffice 
 almost completely to remove the radium from the barium. 
 
 The discovery of radium gave a great impetus to the chemical 
 examination of radioactive minerals in order to see if other radio- 
 active substances were present. Debierne 2 succeeded in ob- 
 taining a new radioactive body called "actinium." Giesel 1 
 independently observed the presence of a new radiating body 
 which he called the "emanating substance," and later "eman- 
 ium," on account of the rapid emission from it of a short-lived 
 radioactive emanation or gas. Recent work has shown that the 
 substances separated by Debierne and Giesel are identical in 
 radioactive properties and must contain the same element. 
 Hofmann and Strauss 3 separated an active substance which was 
 precipitated with lead and called by them "radiolead," while 
 Marckwald 4 later obtained from pitchblende residues some ex- 
 tremely active matter which he named " radiotellurium, ' ' since 
 it was initially separated with tellurium as an impurity. 
 
 None of these active bodies except radium have yet been ob- 
 tained in a pure state. We shall see later that the active ele- 
 ment present in the radiotellurium of Marckwald is almost 
 certainly identical with that in the polonium of Mme. Curie; 
 it will also be shown that the active elements present in radio- 
 lead and radiotellurium are in reality produced from the radium 
 present in pitchblende, or, in other words, both are products 
 of the transformation of the radium atom. 
 
 The possibility of using very active preparations of radium 
 as a source of radiation led to a close examination of the nature 
 of the rays emitted so freely from this substance. Giesel 6 ob- 
 served in 1899 that the more penetrating rays, known as /3 rays, 
 
 1 Giesel: Ber. d. d. chem. Ges., p. 3608, 1902 ; p. 342, 1903. 
 
 2 Debierne: Comptes rendus, cxxix, p. 593 (1899) ; cxxx, p. 206 (1900). 
 8 Hofmann and Strauss: Ber. d. d. chem. Ges., p. 3035, 1901. 
 
 * Marckwald : Ibid., p. 2662, 1903. 
 
 5 Giesel: Annal. d. Phys., Ixix, p. 834 (1899). 
 
10 RADIOACTIVE TRANSFORMATIONS 
 
 were deflected in a magnetic field in the same direction as the 
 cathode rays, indicating that they consisted of negatively charged 
 particles projected with great speed from the active substance. 
 
 This was substantiated by experiments of Becquerel, 1 who 
 examined the deviation of a pencil of rays both in a magnetic and 
 electric field. His results showed that the /3 particles had the 
 same small mass as the particles of the cathode stream, whose 
 corpuscular nature had previously been demonstrated by J. J. 
 Thomson. The ft particle was in fact identical with the elec- 
 tron set free by the electric discharge in a vacuum tube. 
 
 The /3 particles were projected from radium at different 
 speeds, but their average velocity was much greater than that 
 impressed on the electron in a vacuum tube, and in many cases 
 approached closely to the velocity of light. This property of 
 radium of emitting a stream of fi particles at different speeds 
 was later utilized by Kaufmann 2 to determine the variation of 
 the mass of the ft particle with velocity. J. J. Thomson had 
 shown in 1887 that a charged body in motion possessed electri- 
 cal mass in virtue of its motion. The theory of this action was 
 later developed by Heaviside, Searle, Abraham, and others. 
 
 The moving charge acts as an electric current, and a magnetic 
 field is generated round the body and moves with it. Magnetic 
 energy is stored in the medium surrounding the charged body, 
 which consequently behaves as if it had a greater apparent mass 
 than when uncharged. This additional electric mass, accord- 
 ing to the theory, should be constant for small speeds but should 
 increase rapidly as the velocity of light is approached. 
 
 Kaufmann found from his experiments that the apparent mass 
 of the electron did increase with speed, and that the increase was 
 rapid as the velocity of light was approached. By comparing 
 the theory with experiment, he concluded that the apparent mass 
 of the ft particle was entirely electrical in origin, and that there 
 was no necessity to assume the presence of a material nucleus 
 on which the charge was distributed. 
 
 This was a very important result, for it indirectly offered a 
 
 1 Becquerel : Comptes rend us, cxxx, p. 809 (1900). 
 
 2 Kaufmann : Physik. Zeit, iv, No. 1 b, p. 54 (1902). 
 
HISTORICAL INTRODUCTION 11 
 
 possible explanation of the origin of mass, which has always 
 been such an enigma to science. If a charge of electricity in 
 motion exactly simulates the properties of mechanical mass, it 
 is possible that the mass of matter in general may be electrical 
 in origin, and may result from the movement of the electrons 
 constituting the molecules of matter. Such a point of view, 
 while most suggestive and important, cannot at present be con- 
 sidered more than a justifiable speculation. 
 
 Villard 1 in 1900 found that radium emitted in addition to a 
 and fi rays, a third type of radiation, now called the 7 rays, 
 which are of an extremely penetrating character. These rays 
 are undeflected by a magnetic or electric field and appear to be 
 a type of penetrating X-rays, which accompany the expulsion of 
 the particles from radium. The presence of these rays was 
 also observed later in thorium, uranium, and actinium. 
 
 In the meantime, the importance of the a rays began to be 
 more clearly recognized. These rays do not possess much power 
 of penetrating matter, for they are stopped in their passage 
 through a few centimetres of air and by a few thicknesses of 
 metal foil. On the other hand, they produce far more ioniza- 
 tion in the gas than the y8 and 7 rays, and the greater proportion 
 of the energy radiated from radioactive bodies is in the form of 
 these rays. They were at first thought to be non-deflectable in 
 a magnetic field, but in 1902 the writer 2 showed that they were 
 deflected to a measurable extent in strong magnetic and electric 
 fields. The direction of deflection was opposite to that of the 
 /3 particles, showing that they carry a positive and not a nega- 
 tive charge of electricity. 
 
 From measurements of the amount of deflection of the rays 
 both in a magnetic and electric field, it was found that the 
 a particle from radium was projected with a velocity about ^ 
 that of light and had a mass about twice that of the hydrogen 
 atom. The a rays from radium thus consist of a stream of atoms 
 of matter projected with great velocity. We shall see later that 
 there is some reason to believe that the a particle is an atom of 
 
 1 Villard: Comptes rendus, cxxx, pp. 1010, 1178 (1900). 
 
 2 Rutherford : Phil. Mag., Feb., 1903; Physik. Zeit., iv, p. 235 (1902). 
 
12 RADIOACTIVE TRANSFORMATIONS 
 
 helium. The main radiations frem radium are thus corpuscular 
 in character, and consist of streams of positively and negatively 
 charged particles. 
 
 In 1903, Sir William Crookes, 1 and Elster and Geitel, 2 inde- 
 pendently observed a very interesting property of the a rays. 
 The a rays from radium or other strongly active substance 
 produce phosphorescence on a screen of crystalline zinc sul- 
 phide (Sidot's blende). On examination of the screen with a 
 lens, the luminosity is found to be not uniform, but to consist 
 of a number of bright points of light, which follow one another 
 in irregular but rapid succession. These "scintillations " are a 
 result, probably indirect, of the bombardment of the screen 
 by the massive a particles, but the exact explanation of this 
 striking phenomenon is still unsettled. 
 
 In the meantime the complexity of the processes occurring in 
 thorium and radium became more evident. The writer 3 in 1900 
 showed that thorium, in addition to the expulsion of a and /3 
 particles, continuously e'mits a radioactive emanation or gas. 
 Both radium and actinium also exhibit a similar property. 
 These emanations consist of gaseous radioactive matter, the 
 radiating power of which rapidly dies away. The emanations 
 of thorium, radium, and actinium can readily be distinguished 
 from one another by the rate at which they lose their activity. 
 The emanations of both actinium and thorium are very short 
 lived, the former losing half of its activity in 3.9 seconds and 
 the latter in 54 seconds. On the other hand, the emanation 
 from radium is far more persistent, and requires a lapse of about 
 four days to reduce the activity to half value. 
 
 About the same time another remarkable action of radium 
 and thorium was disclosed. M. and Mme. Curie 4 found that 
 all bodies placed in the neighborhood of radium salts became 
 temporarily active. A similar property was independently ob- 
 served by the writer for thorium. 5 This property of radium and 
 
 1 Crookes : Proc. Hoy. Soc., Ixxxi, p. 405 (1903). 
 
 2 Elster and Geitel : Physik. Zeit., No. 15, p. 437, 1903. 
 8 Rutherford : Phil. Mag., Jan. and Feb., 1900. 
 
 4 M. and Mme. Curie: Comptes rendus, cxxix, p. 714 (1899). 
 6 Rutherford : Phil. Mag., Jan. and Feb., 1900. 
 
HISTORICAL INTRODUCTION 13 
 
 thorium of "exciting" or "inducing" activity in substances 
 placed near them is directly due to the emanations from these 
 bodies. The emanation is an unstable substance and is trans- 
 formed into a non-gaseous type of matter which is deposited on 
 the surface of all bodies in its neighborhood. 
 
 Another striking property of radium was observed by P. Curie 
 and Laborde in 1903. l A radium compound continuously emits 
 heat at a rate sufficient to melt more than its own weight of 
 ice per hour. In consequence of this a mass of radium always 
 keeps itself at a higher temperature than the surrounding air. 
 This rapid emission of heat by radium is directly connected 
 with its radioactive properties, and will be shown later to result 
 mainly from the bombardment of the radium by the a particles 
 projected from its own mass. 
 
 From the above brief review of the more important proper- 
 ties exhibited by the radioactive bodies, it will be seen that the 
 processes taking place in a mass of radioactive matter are very 
 complicated. In a compound of radium, for example, there 
 occurs a rapid expulsion of a and @ particles accompanied with 
 the generation of 7 rays, a rapid emission of heat, the continuous 
 production of an emanation or gas, and the formation of an active 
 deposit which gives rise to " excited " activity. 
 
 A great advance in the clear understanding of the connection 
 between these various processes resulted from the discovery by 
 Rutherford and Soddy 2 that a very active substance, called 
 thorium X, could be separated from thorium by a simple chemi- 
 cal operation. This thorium X was found to lose its activity 
 in time, while the thorium freed from ThX spontaneously pro- 
 duced a new supply of ThX. In a mass of thorium in radio- 
 active equilibrium, the two processes of growth and change of 
 ThX proceed simultaneously, and the amount of ThX present 
 reaches a constant value when its rate of production from the 
 thorium balances its own rate of change. It was found that 
 the thorium " emanation ' ' was directly produced by ThX, while 
 
 1 P. Curie and Laborde : Comptes rendus, cxxxvi, p. 673 (1903). 
 
 2 Rutherford and Soddy : Phil. Mag., Sept. and Nov., 1902 ; Trans. Chem. Soc., 
 Ixxxi, pp. 321 and 837 (1902). 
 
14 RADIOACTIVE TRANSFORMATIONS 
 
 the emanation in turn gave rise to the active deposit which 
 causes the phenomenon of excited activity. 
 
 Now it has been pointed out that the radioactive property is 
 atomic, and consequently must result from a process occurring 
 in the atom and not in the molecule. In order to explain the 
 results observed, Rutherford and Soddy advanced a theory, 
 known as the " disintegration theory." It is supposed that the 
 atoms of the radioactive bodies are unstable, and that a certain 
 fixed proportion of them become unstable every second and 
 break up with explosive violence, accompanied in general by 
 the expulsion of an a or /3 particle or both together. The 
 residue of the atom, in consequence of the loss of an a particle, 
 is lighter than before, and becomes the atom of a new substance 
 quite distinct in chemical and physical properties from its parent. 
 In thorium, for example, it is supposed that the atom of ThX 
 consists of the thorium atom minus an a particle. Thorium X 
 is unstable and breaks up at a definite rate with the expulsion 
 of another a particle. The residue of the atom of ThX in turn 
 becomes the atom of the emanation, and this in turn breaks up 
 through a further succession of stages. 
 
 The theory was found to account in a satisfactory way for 
 the processes occurring not only in thorium but in all the radio- 
 active bodies. On this view, the radioactive substances are 
 undergoing spontaneous transformation with the appearance of a 
 number of new kinds of matter which are unstable and have 
 a limited life. The radiations accompany the transformations 
 and are produced as a result of an explosive disturbance within 
 the atom. 
 
 The long continued emission of energy from the radioactive 
 bodies does not on this view present any fundamental difficulty 
 and is in accordance with the principle of the conservation of 
 energy. The matter loses in atomic energy at each stage of the 
 transformation, and the energy radiated is derived from the in- 
 ternal energy resident in the atoms themselves. The atom 
 is supposed to consist of a number of charged parts in rapid 
 oscillatory or orbital motion and consequently contains a great 
 store of energy. Part of this energy is kinetic and part poten- 
 
HISTORICAL INTRODUCTION 15 
 
 tial, resulting from the condensation of the electrical charges 
 within the minute volume of the atom. This latent energy of 
 the atom does not ordinarily manifest itself, since the chemical 
 and physical forces at our disposal do not allow us to break up 
 the atom. Part of this energy is, however, released in radio- 
 active changes where the atom itself suffers disruption with the 
 expulsion of one of its charged parts with great velocity. 
 
 This theory has proved of the greatest service in correlating 
 the various phenomena shown by the active substances. In 
 many cases, it offers a quantitative as well as a qualitative 
 explanation of the experimental facts, and has proved of great 
 value in suggesting new lines of attack. 
 
 In addition to its aid in tracing the succession of transforma- 
 tions which occur in the radioelements, this theory has been 
 instrumental in showing that radium is produced from uranium 
 and that the active constituents in radiolead and radiotellurium 
 result from the transformation of radium. 
 
 The application of this theory to the unravelling of the com- 
 plicated series of transformations in radium, thorium, and ac- 
 tinium will form the main subject matter of this treatise. 
 
 The disintegration theory received a strong measure of sup- 
 port from the remarkable observation of Ramsay and Soddy l 
 that the rare gas helium was produced from the radium emana- 
 tion. This in itself supplied unequivocal evidence that there 
 was an actual transformation of matter taking place in radium, 
 one of the products of which was the inactive gas helium. 
 
 It will be seen later that the weight of evidence points to the 
 conclusion that the a particle from radium is an atom of helium. 
 On this view helium arises during the transformation of each 
 product which emits a rays. Such a conclusion, apart from 
 other evidence, is also supported by the recent observation of 
 Debierne that helium is produced from actinium as well as 
 from radium. 
 
 In the above review we have traced the main line of advance 
 of knowledge in the field of radioactivity, but there has been a 
 rapid and important advance in another direction. 
 
 1 Ramsay and Soddy : Proc. Koy. Soc., Ixxii, p. 204 (1903) ; Ixxiii, p. 341 (1904). 
 
16 RADIOACTIVE TRANSFORMATIONS 
 
 Elster and Geitel 1 showed in 1901 that radioactive matter 
 existed in the atmosphere. Later work has shown that the 
 radioactivity of the atmosphere is mainly due to the presence 
 of the radium emanation which diffuses into the atmosphere 
 from the earth. Elster and Geitel and others have made an 
 extensive examination of the radioactivity of soils, and of well 
 and spring waters, and have shown that there is a very wide 
 diffusion of small quantities of radioactive matter throughout 
 the crust of the earth and in the atmosphere. A number of 
 investigators have entered this new field of inquiry, and a large 
 amount of valuable data has already been accumulated. 
 
 While the radioactive elements exhibit the quality of radio- 
 activity in a very marked degree, there is an increasing body of 
 evidence that ordinary matter also possesses this property to a 
 very minute extent, and that the activity observed cannot be 
 ascribed to the presence of traces of the known radioelements. 
 This detection of the radioactivity of ordinary matter has been 
 made possible by the extraordinary delicacy of the electrical test 
 for the presence of radiations which are able to ionize a gas. 
 
 When it is remembered that the initial discovery of the radio- 
 activity of uranium was made in 1896, and that the first evi- 
 dence of the presence of radium was obtained in 1898, it will 
 be seen how rapid has been the' advance in our knowledge of 
 this complicated subject. A very large mass of experimental 
 facts has now been accumulated, and their connection with one 
 another has been made clear by the adoption of a simple theory. 
 The rapidity of this advance has seldom, if ever, been equalled 
 in the history of science, and it is of interest to examine the 
 influences that have made it possible. 
 
 It is not due to the number of workers in the field, for until 
 the last year or two the subjept has been represented by com- 
 paratively few investigators. The main reason for the rapidity 
 of the advance lies in the remarkably opportune time at which 
 the new field was opened up, and the influence upon it of the 
 rapid extension of our knowledge of the passage of electricity 
 through gases. 
 
 l Elster and Geitel : Physik. Zeit., ii, p. 590 (1901). 
 
HISTORICAL INTRODUCTION 17 
 
 In this connection it is of interest to note that the discovery 
 of the radioactive property of uranium might accidentally have 
 been made a century ago, for all that was required was the ex- 
 posure of a uranium compound on the charged plate of a gold- 
 leaf electroscope. Indications of the existence of the element 
 uranium were given by Klaproth in 1789, and the discharging 
 property of this substance could not fail to have been noted if 
 it had been placed near a charged electroscope. It would riot 
 have been difficult to deduce that the uranium gave out a type 
 of radiation capable of passing through metals opaque to ordi- 
 nary light. The advance would probably have ended there, for 
 the knowledge at that time of the connection between electricity 
 and matter was far too meagre for an isolated property of this 
 kind to have attracted much attention. 
 
 It is not necessary, however, to go so far back to illustrate 
 the great influence that the electrical discoveries in the allied 
 field of discharge of electricity through gases have had on the 
 rapid development of radioactivity. If the discovery had been 
 made even a decade earlier, the advance must necessarily have 
 been much slower and more cautious. At that time the possi- 
 bility of the existence of radiations capable of penetrating 
 matter opaque to light had not even been considered, and the 
 true nature of the cathode rays was still a matter for conjec- 
 ture. The character of the radiations from radioactive matter, 
 as we know them to-day, could only have been deduced after 
 a long and laborious series of researches, for not only would the 
 experimenter have had no guidance from analogy, but he must 
 of necessity have developed the methods of attack de novo under 
 difficult conditions. It would have been necessary, also, to have 
 examined in detail the nature of the discharging action of the 
 rays, for on this is based the most important method of measure- 
 ment in radioactivity. 
 
 Let us now examine the conditions that existed during the 
 actual development of the subject. We have seen that the 
 mechanism of the conduction of electricity through gases had 
 been developed primarily by a study of the conductivity of 
 gases exposed to X-rays, and of the discharge of electricity in 
 
 2 
 
18 RADIOACTIVE TRANSFORMATIONS 
 
 a vacuum tube. The knowledge thus obtained was directly 
 applied to the corresponding ionization produced by the radia- 
 tions from active matter and served as a foundation for the elec- 
 tric method of measurement which has been utilized as a rapid 
 quantitative means of radioactive analysis. When the fi rays 
 of radium were found to be deflected in a magnetic field in the 
 same way as the cathode rays, it was only necessary, in order 
 to prove their identity, to adopt the methods which had been 
 familiar to science for several years. In a similar way, the 
 behavior of the non-deflectable 7 rays was directly compared 
 with the known properties of X-rays, while the a rays were 
 found to be analogous in some respects to the canal rays of 
 Goldstein, which had been shown previously by Wien to be 
 deflected by a magnetic and electric field. 
 
 The influence of the ionization theory on the development of 
 radioactivity has been equally marked in other directions. The 
 determination of the charge carried by an ion has been of the 
 greatest utility in determining the order of magnitude of the proc- 
 esses occurring in radioactive matter. These data have been 
 of great value in determining the number of a and /3 particles 
 emitted by radium and in suggesting the probable amount of 
 emanation and of helium liberated from it. Such calculations 
 have enabled us to fix with some certainty the rate at which 
 radium and the other radioactive bodies suffer disintegration, 
 and also to determine beforehand the magnitude of many physi- 
 cal and chemical quantities, and have thus indirectly suggested 
 the methods of attack necessary to solve the various problems 
 which have arisen. 
 
 The fortunate combination of events in the history of radio- 
 activity is strikingly illustrated by the discovery that helium is 
 evolved by the radium emanation. This rare gas has a dramatic 
 history, for its presence was first observed in the sun by Janssen 
 and Lockyer in 1868; but it was not until 1895 that its pres- 
 ence in the rare mineral clevite was observed by Ramsay. An 
 examination of its physical and chemical properties had hardly 
 been completed when Ramsay and Soddy, guided by the disin- 
 tegration theory, made an examination of the gases liberated 
 
HISTORICAL INTRODUCTION 19 
 
 from radium, and discovered that helium was a product of the 
 transformation of radium. If helium had not a short time 
 before been found in radioactive minerals, it is safe to say that 
 this most striking property of radium of producing helium 
 would have long remained hidden. 
 
 While the ionization theory has played a prominent part in 
 the extension of our knowledge in radioactivity, the benefits 
 have not been altogether one-sided, for the results obtained 
 from a study of radioactivity have greatly assisted in extending 
 and confirming the ionization theory. It has placed in the hands 
 of the experimenter a constant and powerful source of ionizing 
 radiation in place of a variable source like X-rays, and this has 
 been of great service in obtaining accurate data. In addition, 
 the results of Kaufmann on the variation of the mass of the /3 
 particle from radium with its velocity, have been an important 
 factor in confirming and extending our conception of electrons. 
 
 Examples of this kind could readily be multiplied, but suffi- 
 cient have been given to illustrate the close connection that has 
 existed and still exists between these two distinct lines of in- 
 vestigation, and the influence which each has exerted upon the 
 development of the other. 
 
 RADIATIONS FROM ACTIVE BODIES 
 
 A brief summary will now be given of the chief properties 
 and nature of the a, /3, and 7 rays emitted from radioactive 
 bodies. All three types of rays possess in common the proper- 
 ties of acting on a photographic plate, of exciting phosphores- 
 cence in certain substances, and of discharging electrified bodies. 
 The three types of rays can be distinguished from one another 
 by their difference in penetrating power and by the effect upon 
 them of a magnetic or electric field. The a rays are completelv 
 stopped by a layer of aluminium about .05 mm. in thickness; 
 the greater part of the (3 rays by 5 mms. of aluminium, while 
 a thickness of at least 50 cms. of aluminium would be required 
 to absorb most of the 7 rays. The relative penetrating power of 
 the three types of rays is thus about in the ratio 1 : 100 : 10000. 
 It must be borne in mind, however, that this is an average value, 
 
20 RADIOACTIVE TRANSFORMATIONS 
 
 for each type of radiation is complex and consists of rays 
 unequally absorbed by matter. 
 
 The a rays consist of positively charged particles projected 
 with a velocity of about twenty thousand miles per second. 
 The apparent mass of the a particle is about twice that of the 
 hydrogen atom. Although the magnetic deflection of these rays 
 has only so far been observed in the case of active substances 
 like radium and polonium, there can be little doubt that the 
 a rays from the other radioactive bodies are of a very similar 
 nature. 
 
 The & rays consist of negatively charged particles projected 
 with great velocities. Their apparent mass is about r oV o of that 
 of the hydrogen atom, and they are identical in all respects 
 except velocity with the cathode-ray particles set free in a 
 vacuum tube. 
 
 In the case of radium, the ft particles are projected with a 
 wide range of velocity, the maximum approaching very closely 
 to the velocity of light, ft rays are also given out by uranium, 
 thorium, and actinium. 
 
 The 7 rays are not deflected by a magnetic or electric field, 
 and closely resemble in general properties the very penetrating 
 X-rays produced in a hard vacuum tube. According to present 
 views, the 7 rays must thus be supposed to be a t} T pe of wave 
 motion in the ether, consisting probably of pulses set up as 
 a result of the expulsion of ft particles. Only those active 
 substances which emit ft rays give rise to 7 rays. The 7 rays 
 are emitted from uranium, thorium, radium, and actinium, but 
 the rays from uranium and actinium have not such a marked 
 power of penetration as the rays from thorium and radium. 
 
 Each of these three primary types of rays falling upon matter 
 gives rise to secondary rays. In the case of the a rays the 
 secondary radiation consists of negatively charged particles 
 (electrons) projected at velocities comparatively small com- 
 pared with those of the^cparticles themselves. The secondary 
 rays arising from the ft and 7 rays consist in part of electrons 
 projected with considerable velocity. These secondary rays in 
 turn produce tertiary rays, and so on. 
 
HISTORICAL INTRODUCTION 
 
 21 
 
 If a strong magnetic field is applied at right angles to a 
 pencil of a, /3, and 7 rays, the three types of rays are sepa- 
 rated from each other. This is shown in Fig. 1, where the 
 magnetic field is acting downwards, perpendicular to the plane 
 of the paper. The ft rays 
 are bent to the right, the a 
 rays to the left, while the 
 7 rays are unaffected. The 
 ft rays consist of particles 
 having unequal velocities 
 and consequently traversing 
 circular orbits of different, 
 radii of curvature. The 
 magnetic deflection of the 
 a rays compared with that 
 of the ft rays is much exag- ' 
 gerated in the figure. FlG - l - 
 
 The relative mass, veloci- Separation of radium rays by the action 
 
 ty,and kinetic energy of the of a ma g netic field - 
 
 average a and ft particles are shown in Fig. 2, where the 
 
 volume of a sphere represents mass and energy, and the length 
 
 of line represents velocity. 
 
 It will be seen from this illustration that although the average 
 
 ft particle has a much higher velocity than the average a parti- 
 cle, its energy of mo- 
 tion on account of its 
 relatively small mass 
 is much less than 
 that of the average 
 ^particle. This re- 
 sult is in accordance 
 with the observed 
 
 MASS VELOCITY ENERGY 
 
 4 
 
 o 
 
 
 
 
 
 P 
 
 
 
 
 ( 
 
 
 
 
 FIG. 2. 
 
 result that the ioni- 
 zation and heating 
 
 effect produced by the a particle are much greater than for the 
 
 ft particle. 
 
 The writer has recently shown that one gram of radium in 
 
22 RADIOACTIVE TRANSFORMATIONS 
 
 radioactive equilibrium emits about 7 x 10 10 ft particles and 
 about 2.5 X 10 n a particles per second. Four a particles are 
 thus expelled from radium for each ft particle. 
 
 RADIOACTIVE SUBSTANCES 
 
 Below is given a list of the radioactive substances which have 
 so far been separated. The nature of the radiations, and the 
 presence or absence of an emanation is also noted. The " period " 
 of the emanation denotes the time required for its activity to 
 fall to half value. 
 
 Uranium : a, (3, and y rays, but no emanation. x 
 
 Thorium : a, ft, and y T&VS ; an emanation, period 54 seconds. 
 
 Radium : a, ft, and y rays ; an emanation, period 3.8 days. 
 
 c mmm ) ^^ g an emana ^i ou period 3.9 seconds. 
 
 Emanium ) 
 
 v only a rays : no emanation. 
 Radiotellurmm ) 
 
 Radiolead (some time after preparation) : a, ft, and y rays, but 
 no emanation. 
 
 These substances, with the exception of polonium, continue 
 to radiate for long periods of time. In addition, there are a 
 number of radioactive products arising from each radioelement 
 which have a comparatively short radioactive life. These prod- 
 ucts are intrinsically as important as the more permanently 
 active substances, and have an equal right to be called elements. 
 On account of the rapidity of their transformation, they exist in 
 extremely small quantity in pitchblende, and are never likely 
 to be obtained in sufficient quantity to be examined by ordinary 
 chemical methods. Polonium and radiotellurium, which con- 
 tain the same radioactive constituent, differ from the other 
 substances in emitting only a rays. In regard to their length of 
 life, they occupy an intermediate position between the rapidly 
 transformed products like the emanation, and a very slowly 
 changing substance like radium. The activity of radiotellurium 
 falls to half value in about 140 days, while the corresponding 
 time for radium is about 1300 years. 
 
HISTORICAL INTRODUCTION 23 
 
 With the exception of uranium, thorium, and radium, none of 
 these substances have been sufficiently purified to determine 
 their atomic weight or spectrum. It seems likely, however, 
 that actinium will prove to be an element at least as active as 
 radium. It will also be shown later that radiotellurium and 
 radiolead in a pure state should be much more active, weight 
 for weight, than radium. 
 
 The activity of a given substance which emits a rays de-> 
 pends on the number of a particles shot out per second, and } 
 this, for equal weights, is inversely proportional to the " period " 
 of that substance. For example, the actinium emanation whose 
 period is 3.9 seconds must be, weight for weight, at least one 
 thousand million times as active as radium. It is on account 
 of their enormous activity and consequent rapidity of transforma- 
 tion that such substances can never be obtained in sufficient 
 quantity for chemical analysis. It is only the more slowly 
 changing substances like radium, radiolead, and radiotellurium 
 that collect in sufficient quantity in pitchblende to be chemi- 
 ally isolated in appreciable quantity. 
 
 It will also be shown later that the radiations emitted from 
 uranium, radium, thorium, and actinium arise only in part from 
 the primary active substance itself. The /3 and 7 rays in all 
 -cases are emitted from the products of the transformation of 
 these elements. These are mixed with the parent substance 
 and add their radiations to it. 
 
 METHODS OF MEASUREMENTS 
 
 There are three general properties of the rays from radio- 
 active substances which have been utilized for the purpose of 
 measurements, depending on (1) the action of the rays on a 
 photographic plate, (2) the phosphorescence excited in certain 
 crystalline substances, (3) the ioriization produced by the rays 
 in a gas. Of these the phosphorescent method is limited to 
 substances like radium, actinium, and polonium which emit very 
 intense radiations. The a, /3, and 7 rays all produce a marked 
 luminosity in the platinocyanides and in the mineral willemite 
 (zinc silicate). The mineral kunzite responds mainly to the ]3 
 
24 RADIOACTIVE TRANSFORMATIONS 
 
 and 7 rays, while Sidot's blende (crystalline zinc sulphide) re- 
 sponds mainly to the a rays. Besides these there are a large 
 number of substances in which a more or less feeble luminosity 
 is excited by the rays. The property of the a rays of producing 
 scintillations on a screen covered with zinc' sulphide is especially 
 interesting, and it has been found possible by this method to de- 
 tect the a rays emitted by feebly active substances like uranium, 
 thorium, and pitchblende. Screens of zinc sulphide have been 
 used as an optical method for demonstrating the presence of the 
 emanations from radium and actinium. Speaking generally, 
 the phosphorescent method, while very interesting as an optical 
 means for examining the rays, is very limited in its application 
 and is only roughly quantitative. 
 
 The photographic method proved of great service in the early 
 development of radioactivity, but has gradually been displaced by 
 the electric method as quantitative determinations have become 
 more and more necessary. It has proved of special utility in 
 examination of the curvature of the path of the rays in magnetic 
 and electric fields. On the other hand, it does not readily lend 
 itself to quantitative comparisons and is very limited in its appli- 
 cation. In the case of feebly active substances like uranium 
 and thorium, long exposures are necessary to produce much 
 photographic effect. It cannot be utilized to follow the rapid 
 changes of activity which are exhibited by many radioactive 
 products, and is not sufficiently sensitive to detect the presence 
 of rays which are readily observed by the electric method. 
 
 The development of the subject of radioactivity has largely 
 depended on the electric method of measurement, which is uni- 
 versally applicable, and far transcends in delicacy either of the 
 other two methods. It readily lends itself to rapid quantitative 
 measurements, and can be applied to all the types of radiation 
 which possess the ionizing property. 
 
 This method is, as we have seen, based on the property of 
 the a, & and y rays of producing charged carriers or ions in the 
 volume of the gas traversed b^ the radiations. Suppose that a 
 layer of radioactive substance urahTtrm^ forexample is placed 
 on the lower of two insulated parallel plates, A and B. (Fig. 3). 
 
HISTORICAL INTRODUCTION 25 
 
 The gas between the plates is ionized at a constant rate by the 
 radiations, and there results a distribution of positive and nega- 
 tive ions in the volume of air. If no electric field is acting, the 
 number of ions does not increase indefinitely, but soon reaches 
 a maximum, when the rate of production of fresh ions by the 
 radiations exactly compensates for the decrease in the number 
 due to the recombination of the positive and negative ions. 
 This latter effect will obviously tend to take place when the 
 positive and negative ions in the course of their movement come 
 within the sphere of one another's attraction. Suppose now 
 that the plate A is kept charged to a constant potential V, and 
 
 B - 
 
 
 e/t*Ttt 
 
 FIG. 3. 
 
 that the rate at which B, initially at zero potential, gains an 
 electric charge is determined by a suitable measuring instru- 
 ment, for example, a quadrant electrometer. 
 
 Under the influence of the electric field, the positive ions 
 travel to the negative plate and the negative ions to the positive. 
 There is consequently a current through the gas, and the plate 
 B and its connections acquire a positive charge. The rate at 
 which the plate B rises in potential is a relative measure of the 
 current through the gas. When V has a small value, the cur- 
 rent is small, but gradually increases with rise of V, until a 
 stage is reached where the current increases very slightly for 
 a large increment of the value of V. The relation between the 
 
26 RADIOACTIVE TRANSFORMATIONS 
 
 current and the applied voltage is seen in Fig. 4. The shape 
 of this curve receives a simple explanation on the ionization 
 theory. The ions move with a velocity proportional to the 
 strength of the electric field. In a weak field there is thus 
 a slow movement of the positive and negative ions past one 
 another. A large proportion of the ions have time to recom- 
 bine before they reach the electrodes, and the current observed 
 through the gas is consequently small. As the voltage in- 
 creases, the velocity of the ions increases, and there is less time 
 for recombination. Finally, in a strong field practically all the 
 
 VOLTS. 
 
 FIG. 4. 
 Typical saturation curve for an ionized gas. 
 
 ions are swept to the electrodes before any appreciable recom- 
 bination can occur. The maximum or "saturation" current 
 through the gas is then a measure of the charge carried by the 
 ions produced per second by the radiation, i. e., it is a measure 
 of the total rate of production of ions. 
 
 The term "saturation," which was applied initially from the 
 resemblance of the current-voltage curve to the magnetization 
 curve for iron, is not very suitable, but has come into use as a 
 convenient though inaccurate method of expressing an experi- 
 mental fact. 
 
 Other conditions being the same, the voltage required to 
 produce saturation increases with the intensity of the ioniza- 
 
VER31TY V 
 
 HISTORICAL INTRODUCTION 27 
 
 tion, i. <?., with increase in activity of the substance under ex- 
 amination. Increase of the distance between the plates lowers 
 the value of the electric field and increases the distance over 
 which the ions move. Both of these conditions tend to increase 
 the voltage to be applied in order to give saturation. 
 
 It is found experimentally that for parallel plates not more 
 than 3 or 4 cms. apart, 300 volts is sufficient to produce ap- 
 proximate saturation, using substances whose activities are not 
 greater than 1000 times that of uranium. For intensely ac- 
 tive substances like radium, in order to produce saturation 
 at all, the plates must be close together and a high voltage 
 applied. 
 
 The essential condition for quantitative comparisons by the 
 electric method depends on the measurement of the saturation 
 current, for this is a measure of the total number of ions pro- 
 duced per second in the volume of gas under consideration. 
 
 The electric method can be used with accuracy to compare 
 the relative activity of substances which emit identically the 
 same rays, but differ only in the intensity of the radiations. It 
 serves, for example, to determine accurately the rate at which 
 simple products like the emanations lose their activities. 
 
 Unless other factors are taken into consideration, the electric 
 method cannot be directly used to- compare the relative inten- 
 sity of different types of radiation. For example, the relative 
 saturation currents produced by the a and ft rays, emitted from 
 a thick layer of uranium, under the conditions shown in Fig. 3, 
 cannot be used as a direct comparison of the intensity of the 
 two types of radiations, for on account of their difference in 
 penetrating power, a much smaller fraction of the total energy 
 of the issuing fi rays is absorbed in producing ions between the 
 plates than in the case of the more easily absorbed a rays. Be- 
 fore such comparisons of ionization currents can be used to . 
 measure the relative amounts of energy of the two types of 
 rays, the relative penetrating and ionizing powers of the types 
 of radiation must be accurately known. The main province of 
 the electric method, however, lies in the determination of the 
 variations in the activity of a body which emits rays all of one 
 
28 
 
 RADIOACTIVE TRANSFORMATIONS 
 
 kind, and in this it has proved of great value, and has yielded 
 results of considerable accuracy. 
 
 A variety of methods have been employed to measure the 
 ionization currents produced by the radiations. If a very active 
 substance is under examination, a sensitive galvanometer may 
 be used to measure the saturation current. With slight modifi- 
 cations, the gold-leaf electroscope has proved an accurate and 
 reliable means of measurement, and has played a prominent 
 
 part in the development 
 of radioactivity. Various 
 types of the instrument 
 have been employed. A 
 simple form which I have 
 found very convenient in 
 comparisons of activities is 
 shown in Fig. 5. The ac- 
 tive material is placed on 
 the lower .plate, A, which 
 is mounted on a slide so 
 that it can be easily moved 
 out to place the radioactive 
 material in position. The 
 upper plate, B, placed about 
 
 FIG. 5. 
 
 Simple electroscope for comparing 
 o-ray activities. 
 
 3 cms. above A, is connected 
 with a rod, R, which is rig- 
 idly supported by a cross 
 rod, TT, on two insulating 
 sulphur rods, SS. The aluminium or gold leaf is connected 
 with the upper part of the rod R. The rod C, when con- 
 nected with a suitable voltage, serves to charge the electroscope 
 system. 
 
 The movement of the gold leaf is observed through glass or 
 mica windows by means of a low -power microscope provided 
 with a micrometer scale in the eyepiece. The lower plate, A, 
 and the external case, PP, are connected to earth. 
 
 By suitably adjusting the length of the gold leaf and the 
 position of the boundaries, the time taken for the gold leaf to 
 
HISTORICAL INTRODUCTION 
 
 29 
 
 pass over a definite number of divisions of the scale in the eye- 
 piece may be made nearly constant over a considerable range. 
 The radioactive material, placed in a metal or other conducting 
 vessel, is put in position. The electroscope is then charged 
 and the time taken for the gold leaf to pass over a fixed part of 
 the scale is observed. This must be corrected for the natural 
 leak of the instrument, which is determined before the introduc- 
 tion of the active material. This natural leak may be due in 
 part to a slight leakage over the sulphur 
 supports, or more generally to a weak 
 activity of the walls of the electroscope. 
 All substances are slightly active, and 
 this activity is often increased by radio- 
 active contamination from the presence 
 of radium and other emanations. Two 
 or three hundred volts is sufficient to 
 charge the electroscope, and this insures 
 saturation over the greater part of the 
 range provided that the active material 
 does not cause the electroscope to lose 
 its charge in less than two or three 
 minutes. 
 
 In this way measurements can be 
 made with rapidity and certainty. An 
 accuracy of one per cent can readily be 
 obtained, and with care the precision of 
 measurement may be made still greater. 
 The great advantages of this type of instrument are its simplic- 
 ity, portability, and comparative ease of construction. Such an 
 instrument, if standardized by a constant source of radiation 
 like uranium, is very suitable for determining the variation of 
 activity of substances, which change very slowly with time. 
 
 A modification of this electroscope, first used by C. T. R. 
 Wilson, can be utilized to -measure extraordinarily minute 
 ionization currents. The construction of the apparatus is seen 
 in Fig. 6. 
 
 A clean metal vessel, preferably of brass, of about one litre 
 
 FIG. 6. 
 
 Electroscope for compar- 
 ing and 7 ray activities 
 and for measurement of very 
 weak activities. 
 
30 RADIOACTIVE TRANSFORMATIONS 
 
 capacity, has a gold leaf, L, attached to a rod, R, insulated by 
 a sulphur or amber bead, S, inside the vessel. This is charged 
 by a movable rod, C, or by a magnetic device. After charg- 
 ing, the upper rod, P, is connected to the case of the instrument 
 and to earth. In special cases, if extremely minute currents 
 are to be measured, the rod P is kept connected with a source of 
 potential slightly greater than the potential of the electroscope 
 system. This insures that there is no leak of the charge across 
 the sulphur support. 
 
 The movement of the gold leaf is observed as before with a 
 microscope having a micrometer eyepiece. The great advan- 
 tage of this instrument lies in the fact that the apparatus can 
 be hermetically closed. The rate of leak observed must then 
 be entirely due to the ionization in the interior of the vessel 
 and be independent of external electrostatic disturbances. 
 
 An instrument of this kind is very useful for comparing 
 y8 and 7 ray activities. For the former, the base of the 
 electroscope is removed, and replaced by a sheet of aluminium, 
 about .1 mm. in thickness, which completely stops the a rays, 
 but allows the /3 rays to pass through with little absorption. 
 For measurements of the 7 rays, the vessel is placed on a lead 
 plate about 5 rnms. thick, and the active material placed beneath. 
 The /8 rays are completely absorbed by this thickness of lead, 
 and the ionization of the vessel is then due to the more penetrat- 
 ing 7 rays. 
 
 The most convenient general method of measurement depends 
 on the use of a quadrant electrometer. A very convenient and 
 useful type of electrometer for radioactive and other work has 
 been designed by Dolezalek. 1 The general construction of the 
 instrument is seen in Fig. 7. 
 
 The four quadrants are mounted on amber or sulphur sup- 
 ports. A very light needle, N, is made out of silvered paper 
 and is suspended by a fine quartz fibre or phosphor-bronze strip. 
 The needle is charged to a potential of 100 to 300 volts. If 
 a quartz suspension is used, this is done by lightly touching the 
 metallic support of the needle with a wire connected with the 
 
 1 Dolezalek: Instmmentenkunde, p. 345, 1901. 
 
HISTORICAL INTRODUCTION 
 
 31 
 
 source of potential. It is often more convenient to use a fine 
 phosphor-bronze suspension. The needle may then be directly 
 connected with one terminal of a battery the other pole of which 
 is earthed, and its potential kept constant. By the use of a fine 
 quartz suspension, the sensibility of the instrument, i. e., the 
 number of millimetre divisions, passed over by the spot of light 
 
 FIG. 7. 
 Dolezalek electrometer. 
 
 on the scale for an application of a difference in potential of one 
 volt between the quadrants, may be made very great. A sen- 
 sibility of 10,000 millimetre divisions per volt is not uncommon. 
 Unless, however, a very small current is to be measured, it is 
 not advisable to work with a sensibility greater than 1000 divi- 
 sions per volt, and 200 divisions is often sufficient. 
 
32 
 
 RADIOACTIVE TRANSFORMATIONS 
 
 The quadrant electrometer is essentially an instrument for 
 measuring the potential of the conductor with which it is 
 connected, but is indirectly used in radioactivity to measure 
 ionization currents. The capacity of the electrometer and its 
 connections remains sensibly constant with the movement of 
 the needle, and the rate at which the spot of light moves over 
 the scale is a measure of the rate of rise of potential of the elec- 
 trometer system. This serves as a measure of the ionization 
 current between the electrodes of the testing vessel. 
 
 The general arrangement for measurement is shown in Fig. 8. 
 The active material is placed on the lower of two parallel in- 
 sulated plates, A and B. The plate A is connected with one 
 
 CLKTRQMfTCR, 
 
 TfaTIN* VCSSfL. 
 
 CAQTH. 
 
 FIG. 8. 
 Method of use of electrometer for comparing activities. 
 
 terminal of a battery of suitable potential, and the plate B is 
 connected through a key, K, with one pair of quadrants of the 
 electrometer. When not in use the key connects the quadrants 
 and its connections to earth. When a measurement is required, 
 the earth connection is quietly but quickly broken by means of 
 some mechanical or electro-magnetic device. The plate B and 
 its connections rise in potential, and this is indicated by the 
 movement of the spot of light over the scale. The time taken 
 for the spot of light to move over a definite distance of the scale 
 is then observed, and the number of divisions passed over per 
 second serves as a comparative measure of the ionization current 
 through the gas. 
 
 When the movement of the spot of light becomes too rapid 
 for accurate observation, an additional capacity in the form of 
 
HISTORICAL INTRODUCTION 33 
 
 an air or mica condenser is added to the electrometer system 
 and the rate of movement is reduced to that required. 
 
 Proceeding in this way, activities can readily be compared 
 over a very wide range. The magnitude of the current that 
 can be measured is only limited by the capacity of the con- 
 densers available and the voltage of the battery, which must 
 be sufficient to produce saturation in the testing vessel. 
 
 This use of electroscopes and electrometers to compare activi- 
 ties thus depends on the rate of angular movement of the mov- 
 ing system. By a suitable arrangement, an electrometer may 
 be used as a direct reading instrument for measuring current in 
 the same way as a galvanometer. 
 
 Suppose that the electrometer system is connected to earth 
 through a very high resistance, which obeys Ohm's law. When 
 the earth connection of the quadrants is broken, the plate B 
 (Fig. 8) rises in potential until the supply of electricity to 
 B exactly compensates for the loss of electricity by discharge 
 through the high resistance. Since the deflection of an elec- 
 trometer needle is proportional to the voltage applied, the spot 
 of light will thus move from rest to a steady position, and this 
 deflection is proportional to the ionizing current in the testing 
 vessel. 
 
 For measurements of this character, the resistance used must 
 be of the order 10 n ohms. The main drawback of the method 
 is the difficulty of obtaining suitable resistances of this charac- 
 ter which are at the same time free from polarization and obey 
 Ohm's law. The principle of this method has been employed in 
 experiments by Dr. Bronson l in the laboratory of the writer. 
 
 Such an arrangement is especially suitable fer following with 
 accuracy rapid variations of activity. The deflection is inde- 
 pendent of the capacity of the electrometer and connections, and 
 measurements can be made quickly and accurately over a wide 
 range. 
 
 Some of the types of testing vessels suitable for comparisons 
 of activity by the electrometer method are shown in Figs. 9 
 and 10. 
 
 1 Bronson : Amer. Journ. Science, July, 1905 ; Phil. Mag., Jan., 1906. 
 
 3 
 
34 
 
 RADIOACTIVE TRANSFORMATIONS 
 
 The active material is placed on the lower of two parallel 
 plates, A and B, in a closed vessel (Fig. 9). The insulated 
 plate, B, is attached by ebonite rods to the case of the apparatus 
 
 U.VJSICCI 
 
 oTneier 
 
 
 \ 
 
 
 
 
 
 
 B 
 
 Active Material 
 \ I A 
 
 1 
 
 N 
 
 I 
 
 J 
 
 ]( Wo Battery 
 
 FIG. 9. 
 
 Parallel plate testing vessel. 
 
 which is connected with earth, so that there can be no direct 
 conduction leak from the battery to B. 
 
 In Fig. 10 is shown a cylindrical testing vessel, B, suitable 
 for comparison of the activities obtained on wires or rods. The 
 
 tarth 
 
 Earth 
 
 FIG. 10. 
 
 Cylindrical testing vessel with guard ring for comparing activities 
 on wires or rods. 
 
 inner active electrode, A, passes through an ebonite cork. 
 The conduction leak across the ebonite is avoided by use of 
 the guard ring principle, a metal cylinder, CO', connected with 
 earth dividing the ebonite cork into two parts. In this case, the 
 ebonite has only to be insulated sufficiently for the small rise 
 
HISTORICAL INTRODUCTION 35 
 
 of potential required to cause a suitable deflection of the elec- 
 trometer needle. The adoption of the guard ring principle is 
 advisable in all cases in order to get rid of possible conduction 
 currents across the surface of the insulator. 
 
 An apparatus of this kind is very suitable for determining 
 the decay curves of the excited activity imparted to cylindrical 
 electrodes, and for determinations of the decay of activity of 
 emanations which are introduced into the cylinder. 
 
 The electric method is an extraordinarily delicate means of 
 detecting the presence of minute quantities of radioactive matter. 
 This can be readily illustrated by a simple experiment. A 
 milligram of radium bromide is taken and dissolved in 100 c.c. 
 of water. After thorough mixing, 1 c.c. of this solution is 
 taken and added to 99 c.c. of water. One c.c. of this last 
 solution thus contains 10~ 7 gram of radium bromide. If this 
 is evaporated on a metal vessel, the activity possessed by this 
 minute quantity of radium suffices to cause an extremely rapid 
 discharge when brought near the cap of an electroscope, such 
 as is shown in Fig. 11. If the radium covered plate is placed on 
 the cap of the electroscope, it is impossible for the leaves to 
 retain their charge more than a few seconds. 
 
 Using an electroscope of small natural leak, the presence of 
 10~ n gram of radium can easily be detected by the increased 
 rate of movement imparted to the gold leaf. 
 
 Extraordinarily minute currents can be measured with ac- 
 curacy in an electroscope of the type shown in Fig. 6. For 
 example, Cooke observed that in a well cleaned brass vessel of 
 about one litre capacity, the fall of potential due to the natural 
 ionization of the air inside the electroscope was about six volts 
 per hour. The capacity of the gold-leaf system was about one 
 electrostatic unit. The current is equal to the capacity multi- 
 plied by the fall of potential per second, i. e., 
 
 1x6 
 
 current = ^-^ = 5.6 x 10~ 6 electrostatic units 
 
 X oUO 
 
 = 1.9 X 10~ 15 amperes. 
 
36 RADIOACTIVE TRANSFORMATIONS 
 
 With special precautions, a rate of discharge of -^ of this 
 amount can be accurately measured. 
 
 The number of ions produced in the electroscope can readily 
 be deduced. J. J. Thomson has shown that an ion carries 
 a charge of 3.4 x 1Q- 10 electrostatic units or 1.13 x 10~ 19 
 coulombs. The number of ions produced per second in the 
 air is thus, 
 
 1.9 x 10- 15 
 
 1.13 x 10~ 19 
 
 = 17000. 
 
 Supposing the ionization to be uniform throughout the volume 
 of air, this corresponds to a production of 17 ions per c.c. 
 per second in the volume of air inside the vessel of one litre 
 capacity. 
 
 It will be shown later that the average a particle is able to 
 produce about one hundred thousand ions in its path before it 
 ceases to ionize the gas. We thus see that the electrical method 
 is capable of detecting the ionization produced by a quantity of 
 active matter which on an average expels one a particle per 
 second ; or, in other words, the electric method serves to detect 
 a transformation of matter which takes place at the rate of one 
 
 atom per second. 
 
 ^ ** \ t 
 For the detection of matter which possesses the radioactive* 
 
 property, the electric method thus far transcends in delicacy the 
 use of the spectroscope. In consequence of this, we are able to 
 detect the presence of an active substance like radium, when 
 it exists in almost infinitesimal amount mixed with inactive 
 matter, and also to determine with fair accuracy the amount 
 present. This test is so extraordinarily delicate that the pres- 
 ence of minute traces of radium has been observed in almost 
 every substance which has been examined. 
 
CHAPTER II 
 RADIOACTIVE CHANGES IN THORIUM 
 
 IN the preceding chapter, a brief survey has been made of the 
 more important properties possessed by the radioactive bodies, 
 and a short description has been given of the various methods 
 of measurement of radioactive quantities. In this arid succeed- 
 ing chapters, we shall analyze in more detail the processes oc- 
 curring in the radiaoctive substances and the theories advanced 
 for their explanation. Although in most of the subsequent 
 chapters we shall discuss the properties of radium as our typical 
 radioelement, yet the changes which occur in radium are so 
 numerous and so complicated that it is advisable to consider a 
 simpler illustration before entering upon the more complex 
 problem. 
 
 For this purpose, we shall first consider the succession of 
 changes taking place in thorium ; for in that substance the analy- 
 sis required is of a much simpler character than for radium 
 itself. In this way we shall be able to concentrate our atten- 
 tion upon the main phenomena under consideration without 
 being too much disturbed by details. When once the general 
 principles involved have been made clear, their application to 
 the more complex problem of radium will not present much 
 Additional difficulty. 
 
 This mode of beginning is also historically interesting, for it 
 was as a result of the examination of the processes occurring in 
 thorium that the disintegration theory of radioactivity, which 
 will form the basis of the explanation of radioactive phenomena, 
 was first outlined. 
 
 Thorium compounds have about the same radioactivity, 
 weight for weight, as the corresponding compounds of uranium, 
 and, like that substance, emit a, /5, and 7 rays. We have seen, 
 however, that thorium differs from uranium in emitting, besides 
 
38 RADIOACTIVE TRANSFORMATIONS 
 
 the three types of rays, an "emanation," or radioactive gas. 
 This property possessed by thorium of emitting a volatile radio- 
 active substance, can be readily illustrated by the simple experi- 
 mental arrangement shown in Fig. 11. 
 
 A glass tube, A, is filled with about 50 grams of thorium 
 oxide, or, still better, thorium hydroxide, which emits the 
 emanation more freely than the oxide. This is connected by 
 a. narrow tube, L, of about a metre length, with a suitable, well 
 insulated electroscope. On charging the electroscope, the leaves 
 converge very slowly as the thorium compound has no appreci- 
 able effect in ionizing the gas inside the electroscope. If, how- 
 ever, a slow steady stream of air from a gas bag or gasometer 
 
 7 
 
 +/ 
 
 FIG. 11. 
 Discharging power of the thorium emanation. 
 
 is passed over the thorium, no effect is observed in the electro- 
 scope for a definite interval, which is a measure of the time 
 taken for the air current to travel from A to the electroscope. 
 The leaves are then seen to collapse rapidly, the rate of move- 
 ment increasing for several minutes. This discharge of the 
 electroscope is due to the ionizing effect of the thorium emana- 
 tion, which has been conveyed with the current of air into the 
 electroscope. On stopping the air current, the rate of collapse 
 of the gold leaves is seen to diminish steadily, falling to half 
 of its value in about one minute, or, to be more accurate, in 54 
 seconds. After about 5 minutes the effect of the emanation is 
 almost inappreciable. The emanation, left to itself, loses its 
 activity according to an exponential law. In the first 54 seconds 
 the activity is reduced to half value ; in twice that time, i. e. in 
 108 seconds, the activity is reduced to one quarter value, and 
 
 
RADIOACTIVE CHANGES IN THORIUM 39 
 
 in 162 seconds to one eighth value, and so on. This rate of 
 decay of the activity of the thorium emanation is its charac- 
 teristic feature, and serves as a definite physical method of dis- 
 tinguishing the thorium emanation from that of radium or of 
 actinium which decay at very different rates. 
 
 Although the amount of emanation liberated from a kilogram 
 of a thorium compound is far too minute to be detected either 
 by its volume or weight, yet the electrical test of its presence 
 is so extraordinarily delicate that its discharging effect can 
 readily be observed with only a few milligrams of material. 
 
 The amount of emanation liberated into the air from a given 
 weight of thorium varies enormously with the different com- 
 pounds. For example, it is emitted freely by thorium hydroxide, 
 but very slightly by thorium nitrate. Rutherford and Soddy 1 
 made a detailed examination of the " emanating power " of 
 thorium compounds, i. e. the amount of emanation given off 
 into the air per second by a given weight of material, and found 
 that it was much affected by physical and chemical conditions. 
 It is increased by the presence of moisture and by a rise of 
 temperature up to a red heat. Lowering of the temperature to 
 80 C diminishes the emanating power considerably in many 
 cases. The emanation is, however, given off freely, and to an 
 equal extent by all compounds of thorium in solution. This is 
 most readily shown by bubbling a current of air through the 
 solution, when part of the emanation escapes mixed with the 
 air. Rutherford and Soddy showed that the great variations 
 in emanating power under different conditions were not due to 
 differences in the rate of production of the emanation in the 
 various cases, but merely to variations in the rate of escape of the 
 emanation into the air. Since the thorium emanation loses a 
 large proportion of its activity in a few minutes, any retarda- 
 tion in its rate of escape through the pores of a solid compound 
 will very materially alter the emanating power. They con- 
 cluded that, for equal weights of the element thorium, all 
 thorium compounds produce an equal quantity of the emana- 
 
 i Rutherford and Soddy : Phil. Mag., Sept., 1902. 
 
40 RADIOACTIVE TRANSFORMATIONS 
 
 tion per second, but that the rate of its escape into the gas 
 depends greatly on physical and chemical conditions. 
 
 We shall now briefly consider the chemical nature of the ema- 
 nation itself, disregarding for the moment the substance from 
 which it arises. Since the emanation is released in insignifi- 
 cant amount, no direct chemical examination can be made, but 
 the conductivity produced in a gas by the emanation offers a 
 very simple method of testing whether its amount is reduced 
 after it has been acted on by various agents. For example, if 
 the conductivity produced in a testing vessel is unaltered after 
 passing the emanation slowly through a platinum tube at a 
 white heat, and this is what has actually been observed, we can 
 safely conclude that the emanation is unaffected by exposure to 
 that temperature. In this way Rutherford and Soddy found 
 that the thorium emanation was not acted on appreciably by 
 any physical or chemical agent. The emanation was exposed 
 to such severe treatment that no gas except one of the inert 
 group of the argon-helium family could have possibly survived 
 the various processes without change in amount. It was there- 
 fore concluded that the emanation is a chemically inert gas, 
 which, in respect to the absence of any definite combining prop- 
 erties, is chemically allied to the argon-helium family. 
 
 The material nature of the emanation was strongly confirmed 
 by the fact that it could be condensed from any gas with which 
 it is mixed, by the action of extreme cold. The thorium emana- 
 tion commences to condense from the' air at a temperature of 
 about 120 C. The emanation is in consequence completely 
 stopped by passing the gas with which it is mixed, slowly 
 through a U tube immersed in liquid air. From the rate of 
 diffusion of the emanation through air and other gases, it has 
 been deduced that the emanation is a gas of heavy molecular 
 weight. 
 
 We have already seen that the emanation loses its activity 
 rapidly according to an exponential law. If / is the initial 
 activity, the activity I t at any time later is given by the 
 equation 
 
RADIOACTIVE CHANGES IN THORIUM 
 
 41 
 
 where X is a constant and e is the base of Naperian logarithms. 
 Since the activity falls to half value in about 54 seconds the 
 value of 
 
 A = logg2 = .0128 (sec)- 1 . 
 54 
 
 The decay curve of the thorium emanation is shown in Fig. 
 12. If the logarithm of the activity at any time is plotted with 
 
 /oo 
 
 60 
 
 /SO tOO 
 
 T/MC IN SCCQf*Q9. 
 
 200 
 
 2*0 
 
 380* 
 
 FIG. 12. 
 Decay of activity of the thorium emanation. 
 
 respect to time, the points all lie in a straight line. This is 
 shown in the figure, the initial value of the logarithm of the 
 activity being taken as 100 for convenience. 
 
 This value of X is a characteristic constant of the emanation 
 and will be termed the "radioactive constant." As far as obser- 
 vation has at present shown, its value is independent of phys- 
 ical or chemical conditions. For example, the value of X is 
 
42 RADIOACTIVE TRANSFORMATIONS 
 
 the same for the emanation when condensed by liquid air at a 
 temperature of 186 C, as under normal conditions. 
 
 All simple radioactive products lose their activity according 
 to an exponential law, and it is convenient to employ a single 
 term to denote the time taken by any simple product to lose 
 half of its activity. The term "period " of a product will be 
 used in this sense to avoid circumlocution. 
 
 We must now consider the interpretation to be placed upon 
 the observed law of decay of activity of the emanation. Is the 
 activity of the emanation merely a transient superficial property, 
 or is it directly connected with some essential change in the 
 emanation itself? Consider for a moment how the activity is 
 measured. The emanation gives out only a rays, which, as we 
 have seen, are heavy positively charged particles projected with 
 a speed of about twenty thousand miles per second. The ioniza- 
 tion observed in the gas is due to the collision of the projected a 
 particles with the gas molecules which lie in their path. The 
 number of ions produced by each a particle under ordinary con- 
 ditions is very great, probably amounting in some cases to about 
 one hundred thousand. The activity measured by the electric 
 method is thus a comparative measure of the number of a par- 
 ticles expelled from the emanation per second. 
 
 The a particles are apparently derived from the atoms of the 
 emanation, and indeed it is difficult to avoid the conclusion 
 that they were not projected initially from rest, but were in 
 a state of rapid motion before their escape from the atom. It 
 can be calculated that the a particle would have to move freely 
 between two points differing in potential by about five million 
 volts in order to acquire its enormous velocity of projection. 
 
 It is difficult to imagine any mechanism either within or out- 
 side the atom capable of suddenly setting in such rapid motion 
 so heavy a mass as that of the a particle. We are almost forced 
 to the conclusion that the a particle was originally in rapid mo- 
 tion within the atom and for some reason suddenly escaped from 
 the atomic system with the velocity it originally possessed in its 
 orbit. We may suppose that the expulsion of an a particle is 
 a result of a violent explosion within the atom. The residual 
 
RADIOACTIVE CHANGES IN THORIUM 43 
 
 atom is lighter than before, and it is to be expected that its 
 physical and chemical properties will be different from those of 
 the parent atom. This, as we shall see later, is observed in 
 all similar cases. As a result of the expulsion of a particles, 
 the emanation of thorium is changed into another distinct sub- 
 stance which behaves like a solid and is deposited on the sur- 
 face of bodies. This product of the decomposition, or rather 
 disintegration, of the emanation will be discussed later. 
 
 If each atom of the emanation on breaking up expels one a 
 particle, the observed law of decay of activity is expressed by 
 the equation 
 
 where n Q is the initial number that breaks up per second, n t is 
 the number at any time t and X is the radioactive constant. 
 The same equation also holds if each atom in disintegrating 
 expels two or more a particles. 
 
 Assuming for the sake of simplicity the probable law that 
 the disintegration of each atom is accompanied by the appear- 
 ance of one a particle, the number N Q of atoms of emanation 
 initially present must be equal to the total number of a particles 
 expelled during the whole life of the emanation. This number 
 is given by 
 
 N Q = f n t dt = n Q f 
 
 t/O t/O 
 
 
 The number TV which remain unchanged after a period t is easily 
 seen to be given by 
 
 - 
 
 Then 
 
 We have thus arrived at the important conclusion that the 
 number of atoms of the emanation which remain unchanged 
 at any time decreases in exactly the same way as the activity, 
 or, in other words, the activity of the emanation is directly 
 
44 RADIOACTIVE TRANSFORMATIONS 
 
 proportional to the number of atoms of emanation present. 
 This is the result naturally to be expected from a priori con- 
 siderations. 
 
 The exponential law of change of radioactive matter is the 
 same as for a so-called monomolecular change in chemistry, 
 observed in special cases when one of the two combining sub- 
 stances is present in a very large proportion compared with the 
 other. The fact that the constant of decay is independent of 
 the concentration of the emanation points to the conclusion that 
 only one changing system is involved. The fact that the con- 
 stant of decay does not depend on the physical and chemical 
 conditions suggests that the changing system is the atom itself 
 and not the molecule. 
 
 The radioactive constant X has a definite physical meaning. 
 We have seen above that 
 
 Differentiating with regard to the time, 
 
 dN 
 
 or, the average number of 'atoms which break up per second is 
 equal to the total number present multiplied by X. The value 
 of X thus represents the fraction of the total number of atoms 
 which disintegrate per second. For the thorium emanation in 
 which half of the emanation breaks up in 54 seconds, the ratio 
 X = .0128 (sec)" 1 . For example, suppose a vessel contained 
 initiallv 10,000 atoms of the thorium emanation. In the first 
 second, on the average, 128 atoms break up per second. After 54 
 seconds, the number of atoms of emanation present is 5000, and 
 the number breaking up per second is 64. After 108 seconds 
 the number of atoms of emanation left is 2500, and 32 bres 
 up per second, and so on. The value of X has thus a definil 
 and important physical meaning for any radioactive product. 
 
 We have spent some time in considering the physical intei 
 pretation to be placed upon the observed law of decay of activ- 
 ity for the emanation, since every radioactive product, so fa 
 
RADIOACTIVE CHANGES IN THORIUM 45 
 
 examined, follows the same law of change, but with a different 
 though definite value of X, which is characteristic for each 
 special type of radioactive matter. The same general explana- 
 tion of the decay of activity may thus be directly applied to any 
 radioactive product. 
 
 EXCITED RADIOACTIVITY OF THOBIUM 
 
 The writer 1 showed that thorium compounds, besides emit- 
 ting three types of rays and an emanation, possessed the follow- 
 
 Thorium Oxide 
 
 Battery 
 
 FIG. 13. 
 Concentration of the excited activity on the negative electrode. 
 
 ing remarkable property. Any body which has been exposed 1 
 in the presence of the thorium emanation becomes itself radio- [ 
 active. This "excited" or "induced" radioactivity, as it has 
 been termed, is not permanent, but decays when the body is re- 
 moved from the presence of the emanation. The activity can 
 be largely concentrated on the negative electrode in a strong 
 electric field. This can readily be shown by exposing a fine 
 wire, AB, (Fig. 13) in the presence of an emanating thorium 
 compound contained in a closed box, V. 
 
 When the wire is the only negatively charged body exposed 
 to the emanation, it becomes intensely active. If it is charged 
 positively, very little activity is observed. A fine wire may in 
 
 1 Rutherford : Phil. Mag., Jan. and Feb., 1900. 
 
46 RADIOACTIVE TRANSFORMATIONS 
 
 this way be made many hundred times more active per unit area 
 than thorium itself. The activity is due to a material deposit 
 on the wire, for it can be partly removed by rubbing the wire 
 with a piece of cloth, and can be dissolved off a platinum wire 
 by strong acids. If the acid is evaporated in a dish to dryness, 
 the active residue is left behind. The active matter can also 
 be driven off by exposing the platinum wire to a temperature 
 above a red heat. This property of the "active deposit," as it 
 will be called, will be discussed in more detail later. The 
 amount of activity under given conditions which can be con- 
 centrated on a body is independent of its chemical nature. 
 Every substance made active in this way behaves as if it were 
 coated with an invisible film of the same radioactive material. 
 Although the active deposit is too small in quantity to be 
 directly observed, the electrical effects produced by it are often 
 large and very readily measured. 
 
 CONNECTION BETWEEN THE ACTIVE DEPOSIT AND THE 
 EMANATION 
 
 The property of producing an active deposit on bodies belongs 
 not to thorium directly, but to the emanation emitted by it. 
 The activity of the deposit, produced by exposure of a body 
 near a thorium compound, depends on the emanating power of 
 the compound. It is much greater, for example, for the hy- 
 droxide than for the nitrate, since the former emits emanation 
 much more freely. If the thorium compound is completely 
 covered by a very thin plate of mica which prevents the escape 
 of the emanation, no excited activity is produced on a body 
 placed outside it. This shows that the excited activity is not 
 a consequence of some action of the rays directly emitted from 
 thorium, since these are only slightly stopped by the mica, but 
 is due to the presence of the emanation. The close connection 
 between the emanation and excited activity is clearly brought 
 out by the following experiment. A slow constant stream of 
 air is passed over a large weight of thorium compound, and the 
 stream of air mixed with emanation is then passed through 
 a long tube, in which cylindrical electrodes, A, B, C, of equal 
 
RADIOACTIVE CHANGES IN ^HORIUM 47 
 
 length, are placed. The arrangement is clearly seen in Fig. 14. 
 The conductivity of the gas, which is a measure of the amount 
 of emanation present, falls off from electrode to electrode, since 
 the emanation loses its activity with time. The ionization cur- 
 rent observed, for example, between the electrode A and the 
 outer cylinder is at first a measure of the amount of emanation 
 present in the space between the cylinder and the electrode. 
 After several hours, the stream of air was stopped, the central 
 rods, A, B, C, were removed, and the excited activity deter- 
 mined by the electric method in an apparatus similar to that 
 shown in Fig. 9. The excited activity was observed to fall off 
 from electrode to electrode in exactly the same proportion as the 
 
 FIG. 14. 
 
 An experiment to show the connection between the emanation and the excited 
 activit}' it produces. 
 
 activity due to the emanation alone. This shows that the ex- 
 cited activity produced is directly proportional to the amount 
 of emanation present; for as the amount of emanation decreases, 
 the excited activity falls off in the same ratio. This experiment 
 also shows conclusively that the emanation causes the excited 
 activity, since the latter is produced at points far removed from 
 the action of the direct radiation from the thorium. The pro- 
 portion that exists between the amount of the active deposit 
 and of the emanation shows clearly that the latter is the parent 
 of the former. It may be supposed that the residue of the atom 
 of the emanation after an expulsion of the a particle becomes 
 the atom of the active deposit. The new atom in some way 
 gains a positive charge, and is conveyed to the negative elec- 
 trode to which it adheres. In the absence of an electric field, 
 
48 RADIOACTIVE TRANSFORMATIONS 
 
 the carriers of the active substance are conveyed by diffusion to 
 the sides of the vessel containing the emanation. The number 
 of particles of the active deposit produced per second in any 
 space should be proportional to the number of particles of 
 emanation which break up per second. The latter number, as 
 we have seen, is proportional to the activity measured by the 
 electrometer. The view that the particles of the active deposit 
 result from the disintegration of the emanation thus leads at 
 once to the conclusion that the amount of excited activity pro- 
 duced in any space is directly proportional to the activity of 
 the emanation present, i. e. to the amount of emanation present. 
 We may thus conclude with confidence that the active deposit 
 is derived from the disintegration of the emanation, or, in other 
 words, that the emanation is the parent of the active deposit. 
 
 There is a striking difference between the physical and 
 chemical properties of the active deposit and of its parent the 
 emanation. The latter, as we have pointed out, is a chemically 
 inert gas, insoluble in acids, which condenses at about 120 C. 
 The former behaves as a solid substance, is soluble in strong 
 acids, and volatilizes at a temperature above a red heat. As a 
 result of the disintegration of the emanation, a new substance 
 is produced which differs completely both in physical and 
 chemical properties from its parent. 
 
 COMPLEXITY OF THE ACTIVE DEPOSIT 
 
 If a body is exposed for several days in the presence of the 
 emanation, the excited activity after removal decays very nearly 
 .according to an exponential law, falling to half value in 11 
 hours. The active deposit emits a, {3, and 7 rays, and the rate 
 of decay is the same whichever type of rays is used as a means 
 of measurement. This at once suggests that a type of radio- 
 active matter may be present which is half transformed in about 
 11 hours, emitting during the process the three types of rays. 
 The active deposit is, however, more complex than this would 
 indicate. If a body is exposed for only a few minutes to a 
 large supply of emanation, the activity after removal varies in 
 .a very different way from that observed after a long exposure 
 
RADIOACTIVE CHANGES IN THORIUM 
 
 49 
 
 to the emanation. The activity is very small at first, but 
 steadily increases for about 3.66 hours, when it reaches a maxi- 
 mum value. After six hours, the activity diminishes accord- 
 ing to the eleven hour period, observed in the case of long 
 exposures. The curve of decay is shown in Fig. 15. 
 
 This variation of the activity appears at first sight very re- 
 markable and difficult to explain. The curve of variation of 
 
 IZO 
 
 10 Q 
 
 60 
 
 20 
 
 40 
 
 /20 
 
 /oo 20 o 
 
 TV/VfC //V f*/MVTCS. 
 
 320 
 
 FIG. 15. 
 
 Variation with time of the excited activity produced after a short exposure 
 to the thorium emanation. 
 
 activity is quite independent of the chemical nature of the body 
 on which the active deposit is obtained. It is shown equally 
 for deposits on thick plates of metal, and on the thinnest sheets 
 of metal foil. The curve, however, can be completely explained 
 if the active deposit contains two distinct substances, one of 
 which is produced from the other. Let us suppose, for the mo- 
 ment, that the emanation changes into a substance called thorium 
 A, and that after deposit on the surface of the body, thorium 
 
50 RADIOACTIVE TRANSFORMATIONS 
 
 A is gradually transformed into another distinct substance, 
 thorium B. The substance thorium A is supposed to be trans- 
 formed without the emission of either a, /3, or 7 rays, but thorium 
 B emits all three types of rays. If the time of exposure to the 
 emanation is very short compared with the period of transforma- 
 tion of thorium A or B, the matter deposited from the emana- 
 tion will at first consist almost entirely of the inactive substance 
 thorium A. The initial activity after removal will in conse- 
 quence be very small. Since A gradually changes into B, and 
 since the transformation of B is accompanied by the emission of 
 rays, the activity will at first increase with the time, for more 
 and more of B is produced. After a definite interval, the rate 
 of transformation of B will exactly compensate for the rate of 
 supply of B due to the change in A. At that moment the ac- 
 tivity will be a maximum, and will afterwards decrease, since 
 the amount of B will steadily diminish. This hypothesis is 
 seen to account in a general way for the shape of the curve, but 
 we shall now show that it also offers a complete quantitative ex- 
 planation. Suppose that the constants of change of A and B 
 are Xj and X 2 respectively, and that n^ atoms of A are deposited 
 on the body during its short exposure to the emanation. After 
 removal of the body, the number P of atoms of A diminishes 
 according to the equation 
 
 P = 
 The rate of change of P is given by 
 
 If Q is the number of atoms of B present at any time t later, the 
 rate of increase of Q is equal to the rate of supply of fresh 
 atoms in consequence of the transformation of A, less the rate 
 of change of the atoms of B itself, i. e., 
 
 (1) 
 
RADIOACTIVE CHANGES IN THORIUM 51 
 
 The solution of this equation is seen to be of the form 
 
 Q = a e-V + b e-W ; 
 and since Q when t = 0, 
 
 a + b = 0. 
 By substitution we find that 
 
 Therefore, 
 
 . 
 
 A 2 A! 
 
 The value of Q increases at first with the time, passes through 
 a maximum, and then diminishes. A maximum is obtained 
 when 
 
 i. e., at a time T given by the equation 
 
 Since B alone gives out rays, the activity It at any time t is 
 always proportional to the amount of B present: i. e., to the 
 value of . We therefore see that 
 
 _* v " ' (2\ 
 
 ^f^ \ T A T> \ y 
 
 t st A* J. rt At 7 . \ / 
 
 where li is the maximum activity. Since the activity, whether 
 for a long or short exposure, finally decays according to an ex- 
 ponential law with the period of 11 hours, it follows, that either 
 thorium A or thorium B is transformed according to this period. 
 Let us suppose for the moment that half of A is transformed in 
 11 hours. The corresponding value of \ is 1.75 x 10~ 5 (sec)~ 1 . 
 It now remains for us to determine the period of B from con- 
 sideration of the experimental curve. Since it is observed that 
 
52 
 
 RADIOACTIVE TRANSFORMATIONS 
 
 the activity reaches a maximum value after a time T = 220 
 minutes, on substituting the values of Xj and T in equation 
 (2), the value of X 2 is found to be 
 
 A 2 = 2.08 X 10~ 4 (sec)- 1 . 
 
 This corresponds to a change in which half of the matter of 
 thorium B is transformed in 55 minutes. On substituting these 
 
 values of Xj, X 2 , and T in equation (2), the value of ~ can 
 
 1 T 
 
 be at once determined. The very close agreement between 
 the values deduced from the theory and the experimental 
 numbers is shown in the following table. 
 
 Time in minutes. 
 
 I t 
 Theoretical values of jf 
 
 l f 
 
 Observed values of -f-. 
 1 T 
 
 15 
 
 .22 
 
 .23 
 
 30 
 
 .38 
 
 .37 
 
 60 
 
 .64 
 
 .63 
 
 120 
 
 .90 
 
 .91 
 
 220 
 
 1.00 
 
 1.00 
 
 305 , 
 
 .97 
 
 .96 
 
 The agreement is equally close for still longer periods. After 
 about 6 hours the activity decreases very nearly exponentially, 
 falling to about half value in 11 hours. 
 
 We thus see that a quantitative explanation of the activity 
 curve can be obtained on the following assumptions : 
 
 (1) That the matter thorium A deposited from the emanation 
 is half transformed in 11 hours, but does not itself emit rays. 
 
 (2) That the matter thorium A changes into thorium B. 
 which is half transformed in 55 minutes, and emits all three 
 types of rays. 
 
 A very interesting point arises in the selection of the periods 
 of the transformation of thorium A and B. We assumed that 
 the period of 11 hours belonged to A rather than to B, but the 
 activity curve itself gives us no information on this question. 
 We see that equation (2) is symmetrical in regard to Xj, X 2 , 
 
RADIOACTIVE CHANGES IN THORIUM 5B 
 
 and in consequence would not be altered by an interchange of 
 their values. In order to settle this question definitely, it is 
 necessary to isolate thorium B from the mixture of A and B, 
 and separately determine its period. If it is found possible to 
 isolate an active product from the mixture of A and B which 
 decays exponentially, falling off to half value in 55 minutes, 
 it follows at once that the "ray " product B has this period, and 
 that the 11 hour period belongs to A, the "rayless" product. 
 This separation has actually been accomplished by several ex- 
 perimental methods, and the results completely confirm the 
 theory already considered, and at the same time illustrate in a 
 remarkable way the differences in physical and chemical proper- 
 ties of the two products, thorium A and B. 
 
 Pegram l examined the radioactivity produced in the elec- 
 trodes by electrolysis of a thorium solution, and, under suitable 
 conditions, obtained a product, the activity of which decayed 
 exponentially, falling to half value in about 1 hour. 
 
 Von Lerch 2 made a number of experiments on the effect of 
 electrolyzing a solution of the active deposit of thorium, ob- 
 tained by solution in hydrochloric acid of the active deposit on 
 a platinum wire. Deposits of varying rates of decay were ob- 
 tained under different conditions, some decaying to half value 
 in 11 hours, and others at a more rapid rate. By using nickel 
 electrodes, he obtained an active substance which decayed ex- 
 ponentially, falling to half value in one hour. Considering the 
 close agreement between the calculated and observed periods, 
 viz., 55 and 60 minutes respectively, there can be no doubt that 
 the ray product, thorium B, had been completely separated 
 from the mixture of A and B by electrolysis. The rates of 
 decay of the deposits obtained under different conditions are 
 readily explained, for in most cases A and B are deposited elec- 
 trolytically together, but in varying proportions. 
 
 This result was still further confirmed by Miss Slater, 3 using 
 a different method. A platinum wire, made active by exposure 
 
 1 Pegram: Phys. Rev., Dec., 1903. 
 
 2 Von Lerch: Annal. d. Phys., Nov., 1903; Akad. Wiss. Wien, March, 1905. 
 
 3 Miss Slater : Phil Mag., May, 1905. 
 
54 RADIOACTIVE TRANSFORMATIONS 
 
 to the emanation, was heated to a high temperature by means 
 of an electric current. Miss Gates had previously l observed 
 that although the activity of a platinum wire was lost by heat- 
 ing to a white heat, yet if the heated body was surrounded by 
 a cold tube, the activity after heating was found to be dis- 
 tributed in undiminished amount over the interior of the tube. 
 This experiment showed that the activity had not been destroyed 
 by the action of a high temperature, but that the active matter 
 had been volatilized by heat and redeposited on the surrounding 
 cold bodies. Miss Slater examined the rates of decay, both of 
 the activity left behind on the wire, and of that distributed on 
 a lead cylinder surrounding the wire, after heating the latter 
 for a short time at different temperatures. After exposure to 
 a temperature of 700 C. for a few minutes, the activity of the 
 wire was slightly reduced. The activity on the lead cylinder 
 was small at first, but increased, reaching a maximum after 
 about 4 hours, and then decaying exponentially with a period of 
 11 hours. This variation of activity is almost exactly the same 
 as that observed (Fig. 15) for a wire exposed for a short in- 
 terval to the thorium emanation; i. e., under conditions in 
 which the matter initially consisted almost entirely of thorium A. 
 This result, then, shows that some thorium A was driven off by 
 heat, and deposited on the surface of the lead cylinder. On 
 heating to about 1000 C. nearly all the thorium A was re- 
 moved, for it was observed that the activity left behind on the 
 wire decayed exponentially, falling to half value in about 1 
 hour. At a temperature of 1200 C., nearly all the thorium B 
 also was volatilized. These results thus show conclusively that 
 the period of the ray product, thorium B, is about 1 hour, and 
 that the period of 11 hours must be ascribed to the rayless 
 product, thorium A. We therefore see that it has been found 
 possible to isolate the components of a mixture of thorium A 
 and B by two distinct methods, the one depending on the dif- 
 ference in electrolytic behavior of the two substances, and the 
 other on their difference in volatility. This is a very interesting 
 result, for it not only indicates the difference in physical and 
 
 1 Miss Gates: Phys. "Rev., p. 300, 1903. 
 
RADIOACTIVE CHANGES IN THORIUM 55 
 
 chemical nature of the two components of the active deposit, but 
 also shows how a separation of two substances existing together 
 in almost infinitesimal amounts can be effected by specially 
 devised methods. 
 
 It is at first sight a most surprising result that we are 
 able, not only to detect the presence, but also to determine the 
 physical and chemical properties of a product like thorium A, 
 which does not manifest its presence by the emission of radia- 
 tions. This, as we have seen, is rendered possible by the fact 
 that the product of its transformation emits rays. But for this 
 property the presence of thorium A or B would never have been 
 detected by the means at our disposal. 
 
 It has been seen that after a long exposure to the thorium 
 emanation, the excited activity at once commences to diminish. 
 This result necessarily follows from general considerations. If 
 the body is exposed in the presence of a constant supply of the 
 emanation for about a week, the activity produced reaches a 
 steady limiting value. When this is the case, the number of 
 atoms of each product supplied per second is equal to the 
 number of each which breaks up per second. Immediately 
 after removal from the emanation, the amount of A begins 
 to diminish according to an exponential law, and it can be 
 shown both theoretically and experimentally that the activity, 
 which is a measure of the amount of thorium B, does not at 
 first diminish accurately according to an exponential law, with 
 an 11 hour period, but somewhat more slowly. Several hours 
 after removal, however, the decay is very nearly exponential. 
 
 It is a matter of interest to observe that the activity for a 
 long exposure does not decay according to the period of the 
 ray-emitting substance, but according to that of the "rayless" 
 product. The decay in such a case will always follow the 
 longer period, no matter whether the substance, which is trans- 
 formed according to that period, gives out rays or not. 
 
 We may at this stage briefly summarize the conclusions 
 arrived at: 
 
 "(1) The thorium emanation is a gas which is half transformed 
 in 54 seconds, and emits only a rays. 
 
56 RADIOACTIVE TRANSFORMATIONS 
 
 (2) The emanation changes into a solid called thorium A, 
 which is half transformed in 11 hours, but which does not emit 
 rays. 
 
 (3) The thorium A in turn changes into a product, thorium 
 B, emitting a, /3, and 7 rays, which is half transformed in about 
 1 hour. 
 
 The successive changes occurring in the emanation are shown 
 diagrammatically below : 
 
 a particle a particle 
 
 A A 
 
 ray 
 Emanation > Thorium A > Thorium B > ? 
 
 At present we have no definite information with regard to the 
 product of transformation of thorium B. It is either inactive, 
 or active to such a minute extent that its properties cannot be 
 determined by the electric method. 
 
 ^N 
 
 SEPARATION OF THORIUM X 
 
 It is now necessary to go back a stage and investigate the 
 origin of the emanation. We shall first consider an important 
 series of experiments by Rutherford and Soddy, 1 which have 
 not only solved this question, but also have thrown a strong 
 light on the processes occurring in thorium. 
 
 A small quantity of thorium nitrate was taken and dissolved 
 in water. Sufficient ammonia was then added to precipitate 
 the thorium present as hydroxide. The filtrate remaining was 
 then evaporated to . dryness, and the ammonium salts driven off 
 by heating. The small residue finally obtained was, weight 
 for weight, over one thousand times as active as the original 
 thorium nitrate. The great activity of this residue, as com- 
 pared with ordinary thorium, can be readily illustrated by means 
 of the electroscope. The active residue obtained from 50 grams 
 of the nitrate causes the gold leaves of the electroscope to collapse 
 in a few seconds, while a weight of thorium nitrate equal to 
 that of the residue causes hardly an appreciable movement. 
 
 1 Rutherford and Soddy : Phil. Mag., Sept. and Nov., 1902. 
 
RADIOACTIVE CHANGES IN THORIUM 57 
 
 The active substance present in this residue was called for 
 convenience, thorium X (ThX). It probably exists in almost 
 infinitesimal quantity mixed with the impurities left behind 
 after the evaporation of the reagent, together possibly with a 
 small trace of thorium which escaped precipitation. Since 
 ThX is derived from the thorium salt, the latter must have 
 been deprived of some of its activity. This was found to be 
 the case, for the thorium hydroxide, so separated, had only 
 about half of the activity to be normally expected. 
 
 The a ray activity of the ThX and the precipitated hydroxide 
 were examined at intervals by means of an electrometer. The 
 activity of ThX was not permanent, but increased for the first 
 day and then decayed exponentially, falling to half value in 
 about 4 days. After a month's interval, the activity sank to 
 a minute fraction of its original value. The curve showing 
 the variation of activity of ThX with time is seen in Fig. 16, 
 Curve I. 
 
 Now let us turn our attention to the precipitated hydroxide. 
 The activity of this decreased to some extent during the first 
 day, passed through a minimum, and then steadily increased 
 again with the time, reaching an almost steady value after 
 a month's interval. These results are shown in Fig. 16, 
 Curve II. 
 
 The two curves of decay of ThX, and of recovery of activity 
 of the thorium, bear a very simple relation to one another. The 
 initial rise in the ThX curve is seen to correspond to a fall in 
 the recovery curve of the thorium, and when the activity of 
 ThX has almost disappeared the activity of the thorium has 
 practically reached a maximum value. The sum of the activi- 
 ties of the ThX and of the thorium from which it was separated 
 is very nearly constant over the whole range of the experiment. 
 The two curves of recovery and decay are complementary to 
 each other. As fast as the ThX loses its activity, the thorium 
 regains it. This relation between the curves is, at first view, 
 most remarkable, and it would appear as if there were some 
 mutual influence between the ThX and the thorium compound 
 from which it was removed, so that the latter absorbed the 
 
58 
 
 RADIOACTIVE TRANSFORMATIONS 
 
 activity lost by the former. This position is, however, quite 
 untenable, for the rise and recovery curves are independent, and 
 are unaltered if the thorium and ThX are kept in sealed vessels 
 far removed from each other. If the thorium hydroxide after 
 
 FIG. 16. 
 Decay of activity of thorium X and recovery of activity of thorium deprived of ThX. 
 
 recovering its activity is again dissolved and ammonia added, 
 the amount of ThX separated is found to be the same as thai 
 obtained from the first experiment. This process can be repeate< 
 indefinitely, and equal quantities of ThX always be separatee 
 
RADIOACTIVE CHANGES IN THORIUM 59 
 
 provided that about a month elapses between each precipitation 
 in order to allow the thorium to regain its lost activity. This 
 shows that there is a fresh growth of ThX in the thorium after 
 each precipitation. 
 
 We shall now consider the explanation of the connection 
 between the decay and recovery curves. For the moment we 
 shall disregard the initial irregularities shown by the two curves, 
 which will be discussed later. If the recovery curve of Fig. 16 
 
 20 
 
 12 16 
 
 Time in Days 
 
 FIG. 17. 
 
 Decay curve of thorium X and recovery curve of thorium, from measurements 
 one day after removal of the thorium X. 
 
 is produced backwards to cut the vertical axis, it does so at 
 a minimum of 25 per cent. The curve of recovery of the lost 
 activity reckoned from this 25 per cent minimum is shown in 
 Fig. 17. In the same figure is shown the decay curve of ThX 
 beginning after the second day and plotted to the same scale. 
 The decay curve of the ThX is exponential, decreasing to half 
 
60 RADIOACTIVE TRANSFORMATIONS 
 
 value in about 4 days. The decrease of activity from the initial 
 value / is given by the equation 
 
 The two curves are complementary, and the sum of the or- 
 dinates at any time is equal to 100 on the arbitrary scale. After 
 4 days, the activity of ThX has decayed to half value, and in 
 the same interval the thorium has regained half its lost activity. 
 The recovery curve is thus expressed by an equation of the form 
 
 where I t is the activity recovered after any time t, and I Q the 
 maximum activity which is regained when a steady state is 
 reached. In this equation X has exactly the same value as for 
 the decay curve. 
 
 Following the same line of argument employed to interpret 
 the decay curve of the emanation (page 43), we may suppose 
 that the ThX is an unstable substance which is half transformed 
 in 4 days, the amount of ThX breaking up per second being 
 always proportional to the amount present. The radiation con- 
 sisting of a rays accompanies the change, and is also proportional 
 to the amount of ThX present. 
 
 Now we have seen that fresh ThX is produced in the thorium 
 after the first supply has been removed. This production of 
 ThX proceeds at a constant rate, but the amount of ThX 
 present in the thorium cannot increase indefinitely, for at the 
 same time the ThX is being changed continuously into another 
 substance. A steady state will obviously be reached when the 
 rate of production of new ThX exactly compensates for the 
 rate of disappearance of ThX due to its own transformation. 
 Now the number of atoms of ThX which break up per second 
 is equal to \N where X is the radioactive constant of ThX, and 
 N is the number of atoms of ThX present at any time. 
 
 A steady state is reached when the number q o atoms of 
 
 
RADIOACTIVE CHANGES IN THORIUM 61 
 
 fresh ThX supplied per second is equal to the number 
 which break up per second, where N is the maximum number 
 present when equilibrium is reached; i. e., 
 
 At any time the rate of increase of the number of atoms 
 
 ctt 
 
 of ThX present is equal to the difference between the rate of 
 supply and the rate of disappearance; i. e., 
 
 dN 
 
 The solution of this equation is of the form 
 
 N= a e~ + b, 
 
 where a and b are constants. Since N = when t = 0, 
 a + b =0, and remembering that when = oo , Nis equal to JV , 
 
 a = b = N , 
 and consequently 
 
 N 
 
 -^ = 1 e - 
 
 ^o 
 
 This theoretical equation expressing the number of atoms 
 of ThX present at any time is thus identical in form with the 
 equation of variation of activity obtained experimentally. We 
 therefore see that the decay and recovery curves of ThX are 
 completely explained on the simple hypotheses : 
 
 (1) That there is a constant production of fresh ThX from 
 the thorium; 
 
 (2) That the ThX is continuously transformed, the amount 
 changing per second being always proportional to the amount 
 present. 
 
 The hypothesis (2) has been previously shown to be merely 
 another method of expression of the observed exponential law 
 of decay of the activity of ThX. 
 
 The first hypothesis can be proved experimentally. The 
 
62 RADIOACTIVE TRANSFORMATIONS 
 
 amount N of ThX present after the growth has been continued 
 for a time t should be given by 
 
 where N Q is the equilibrium amount. 
 
 Since the ThX is half transformed in 4 days, \ = .173 (day)" 1 . 
 At the end of 1 day after complete removal of the ThX, the 
 amount formed consequently should be 16 per cent of the maxi- 
 mum; after 4 days, 50 per cent; after 8 days, 75 per cent, and 
 so on. Now it was found experimentally that three rapid pre- 
 cipitations of thorium by ammonia almost completely freed it 
 from ThX for the time being. After standing for definite 
 periods, the ThX present was removed and the amounts ob- 
 tained were found to be in good agreement with the theory. 
 
 We thus see that the apparent constant radioactivity of 
 thorium is really the result of two opposing processes of growth 
 and decay ; for radioactive matter is being continuously formed, 
 and this matter in turn is continuously changing, and conse- 
 quently losing its activity. There is thus a type of chemical 
 equilibrium in which the rate of production of new matter 
 balances the rate at which the new matter is transformed. 
 
 SOURCE OF THE THORIUM EMANATION 
 
 A thorium compound completely freed from ThX gives off 
 very little emanation, even in a state of solution. On the other 
 hand, the ammonia solution which contains the ThX gives off 
 a large amount. The removal of ThX is thus accompanied by 
 the removal of the emanating power of thorium. It seems prob- 
 able, therefore, that the emanation is derived from ThX, and 
 further experiment has proved this to be the case. If a solution 
 of ThX is taken, and a constant stream of air bubbled through 
 it, the amount of emanation liberated is found to decrease expo- 
 nentially, falling to half value in 4 days. This is exactly the 
 result to be expected if ThX is the parent of the emanation, 
 for the activity of ThX is a measure of the number of atoms of 
 ThX breaking up per second, i. e., a measure of the number 
 
RADIOACTIVE CHANGES IN THORIUM 6B 
 
 of atoms of the new substance which is formed. The rate of 
 production of emanation by the ThX should, on this view, be 
 always proportional to the activity of ThX, and consequently 
 should diminish at the same rate and according to the same 
 law. This, as we have seen, has been experimentally observed. 
 Although the thorium after removal of ThX is for the time 
 almost entirely deprived of the power of emitting an emanation, 
 this property is gradually regained, according to the same law 
 as the recovery law of ThX shown in Fig. 17. This result 
 follows naturally if ThX is the parent of the emanation. The 
 emanating power should be proportional to the amount of ThX 
 present, and should consequently vary pari passu with it. 
 
 We may thus conclude with confidence that the property of 
 emitting an emanation is not a direct property of thorium itself, 
 but belongs to its product ThX. 
 
 INITIAL IRREGULARITIES IN THE DECAY AND 
 RECOVERY CURVES 
 
 We are now in a position to explain the initial irregularities 
 in the decay and recovery curves shown in Fig. 16. The 
 activity of the separated ThX at first increases, while the 
 activity of the precipitated thorium at first diminishes. Now 
 the active deposit produced from the emanation is insoluble in 
 ammonia, and consequently is left behind with the thorium. 
 The ThX after separation produces the emanation, and this, in 
 turn, is transformed into thorium A and B. The activity sup- 
 plied by thorium B more than compensates at first for the decay 
 of the activity of ThX alone. The activity consequently rises, 
 but since the rates of transformation of A and B are rapid com- 
 pared with that of ThX, after about one day equilibrium is 
 practically reached, when very nearly the same number of atoms 
 of ThX, and of each of its products, break up per second. 
 When this is the case the activity of the emanation and of 
 thorium B will vary exactly in the same way as that of the 
 parent substance ThX. The activity of the active residue 
 which is a measure of the activity due to ThX, the emanation, 
 
64 RADIOACTIVE TRANSFORMATIONS 
 
 and thorium B together will in consequence decrease expo- 
 nentially, falling to half value in four days. 
 
 Since the active deposit produced by the emanation in the 
 mass of the thorium compound is not removed with the ThX, 
 the activity due to it must at first diminish, for, in the ab- 
 sence of ThX and the emanation, there is no fresh supply 
 of thorium A and B to compensate for their transformation. 
 The activity of the thorium will thus diminish until the fresh 
 supply of activity due to ThX and its succeeding products com- 
 pensates for the decrease in the activity of the deposit. The 
 activity will then be at a minimum, and will afterwards increase 
 with the time, in consequence of the continued production 
 of ThX. 
 
 The complementary character of the curves of decay and 
 recovery, quite apart from the special considerations here ad- 
 vanced, is a necessary consequence of the laws governing 
 radioactive changes. The rate of transformation, so far as 
 observation has gone, is not affected by physical and chemical 
 conditions. The transformation of ThX, when mixed with 
 thorium, takes place at the same rate, and according to the 
 same laws, as when it is isolated from the thorium by a chemical 
 process. When the activity of a thorium compound has reached 
 a constant value, the activity is then due to the various active 
 products formed in it. If a product is separated by chemical 
 or other means from the thorium compound, the activity due 
 to this product plus that due to the thorium and the active 
 products left behind, must be equal to the constant value of 
 the activity of the original thorium in equilibrium. This follows 
 at once, for otherwise there would be a creation or destruction 
 of radioactivity by the mere removal of one of the products, and 
 this would involve a gain or loss of radioactive energy. If, as in 
 the case of ThX, the separated product first increases and then 
 decreases in activity, there must be a corresponding decrease 
 followed by an increase in the activity of the thorium from 
 which it has been separated, in order that the sum of the activ- 
 ities of the two may be constant. 
 
 This principle of the conservation of the total amount of 
 
RADIOACTIVE CHANGES IN THORIUM 65 
 
 radioactivity applies not only to thorium, but to radioactive 
 substances generally. The total radioactivity of any substance 
 in equilibrium cannot be altered by any physical or chemical 
 agency, although the radioactivity may be manifested in a series 
 of products, capable of separation from the parent substance. 
 There is reason to believe, however, that the radioactivity of 
 the primary active substances is not strictly permanent, but 
 diminishes slowly, although in the case of feebly active elements 
 like uranium and thorium, probably no appreciable change 
 would be detected in a million years. 
 
 With an intensely active body like radium, it will be shown 
 later that in all probability the sum total of the activity will 
 ultimately decay exponentially, decreasing to half value in about 
 thirteen hundred years. Provided, however, that the period 
 of observation is small compared with the life of the primary 
 substance, the principle of the constancy of the radioactivity 
 is a sufficiently accurate expression of the experimental results. 
 Many examples in support of this principle will be found in 
 succeeding chapters of this book. 
 
 METHODS OF SEPARATION OF THORIUM PRODUCTS 
 
 In addition to ammonia, several reagents have been found 
 capable of removing ThX from thorium solutions. Schlundt 
 and R. B. Moore 1 found that pyridine and fumaric acid separate 
 ThX from thorium nitrate solutions. These reagents differ 
 from ammonia in removing the inactive product, thorium A, 
 with the ThX, while the active product, thorium B, is left 
 behind with the thorium. 
 
 Von Lerch 2 has shown that ThX can be separated by elec- 
 trolysis from an alkaline solution of ThX, using amalgamated 
 zinc, copper, mercury, or platinum as electrodes. The period 
 of ThX has been accurately determined, and found to be 3.64 
 days. In addition, Von Lerch found that ThX was deposited 
 on different metals by leaving them for several hours in an 
 alkaline solution of ThX. Iron and zinc removed the greatest 
 
 1 Schlundt and R. B. Moore: Journ. Phys. Chem., Nov., 1905. 
 
 2 Von Lerch: Wien. Ber., March, 1905. 
 
 5 
 
66 RADIOACTIVE TRANSFORMATIONS 
 
 quantity. A nickel plate dipped into an acid solution of the 
 active deposit becomes coated with thorium B, for the activity 
 observed on the metal decays exponentially with a period of 
 1 hour. Other metals similarly treated also became active, but 
 their rates of decay show that they are coated with a mixture 
 of thorium A and of thorium B. 
 
 These results have shown in a striking way the differences 
 in physical and chemical properties of the various thorium prod- 
 ucts. The methods of separation of the infinitesimal quan- 
 tities of matter present are as definite as the ordinary chemical 
 methods, applied to matter existing in considerable amount, 
 while the radiating property serves as a simple and reliable 
 method of qualitative and quantitative analysis. 
 
 CHANGES IN THORIUM 
 
 We have so far shown that thorium produces ThX, and that 
 the latter is transformed into the emanation, which undergoes 
 two further changes into thorium A and thorium B. 
 
 If thorium is subjected to a succession of precipitations with 
 ammonia, extending over several days, the ThX is removed as 
 fast as it is formed, and the active deposit has time to disap- 
 pear. The activity of the thorium then sinks to a minimum 
 of 25 per cent of its value when in equilibrium. The recovery 
 curve of the thorium treated in this way does not show the 
 initial decrease already referred to, but rises steadily, according 
 to the recovery curve shown in Fig. 17. It is thus seen that 
 thorium itself supplies only 25 per cent of the total a ray activ- 
 ity of thorium when in equilibrium, and that the rest is due to 
 ThX, the emanation, and thorium B. Each of these a ray prod- 
 ucts supplies about 25 per cent of the total activity. Such a 
 result is to be expected, for, when in equilibrium, an equal 
 number of atoms of thorium ThX, the emanation, and thorium 
 B must break up per second. This is based on the reasonable 
 supposition, that each atom in breaking up gives rise to one 
 atom of the succeeding product. The results, so far obtained, 
 are completely explained on the disintegration theory put for- 
 ward by Rutherford and Soddy. On this theory, a minute con- 
 
RADIOACTIVE CHANGES IN THORIUM 
 
 67 
 
 slant fraction of the atoms of thorium becomes unstable every 
 second, and breaks up with the expulsion of an a particle. The 
 residue of the atom after the loss of an a particle becomes an 
 atom of a new substance, thorium X. This is far more unstable 
 than the thorium itself, and breaks up with the expulsion of an 
 a particle, half the matter being transformed in 4 days. ThX in 
 turn changes into the emanation, which again breaks up into the 
 active deposit, consisting of two successive products, thorium A 
 and thorium B. The atom of thorium B breaks up with the ac- 
 companiment of an a and a /3 particle, and 7 rays. Thorium A 
 is transformed into thorium B without the appearance of rays. 
 Such a change may consist either of a rearrangement of the parts 
 constituting the atom without the projection of a part of its 
 mass, or of the expulsion of an a particle at too low a velocity 
 to ionize the gas. From the considerations advanced later in 
 Chapter X, the latter supposition does not appear improbable. 
 
 A table of the products of thorium and some of their charac- 
 teristic physical and chemical properties is given below. 
 
 TABLE OF TRANSFORMATION PRODUCTS OF THORIUM. 
 
 Radioactive 
 product. 
 
 Time to be half 
 transformed. 
 
 Nature of 
 rays. 
 
 Some physical and chemical properties. 
 
 Thorium 
 
 About 10 9 years 
 
 a 
 
 Insoluble in ammonia. 
 
 Thorium X 
 
 4 days 
 
 a 
 
 Separated from thorium by its 
 
 
 
 
 
 solubility in ammonia and in 
 
 
 
 
 
 water and by electrolysis ; sep- 
 
 
 
 
 
 arated also by f umaric acid and 
 
 > 
 
 
 
 
 
 pyridine. 
 
 Emanation 
 
 54 sees. 
 
 a 
 
 A chemically inert gas of high 
 
 
 
 
 
 molecular weight ; condenses 
 
 
 
 
 
 from gases at a temperature of 
 
 > 
 
 f 
 
 
 
 -120 C. 
 
 Thorii 
 
 N 
 
 im A 
 
 r 
 
 11 hours 
 
 No rays'l 
 
 Deposited on the surface of bod- 
 ies ; concentrated on the nega- 
 tive electrode in an electric 
 
 Thorium B 
 
 1 hour 
 
 " & 7 ) 
 
 field ; soluble in strong acids ; 
 
 
 
 
 
 volatilized at high tempera- 
 
 
 
 
 
 tures. A is more volatile than 
 
 
 
 
 
 B. A can be separated from B 
 
 
 
 
 
 by electrolysis and by its dif- 
 
 \ 
 
 ' 
 
 r 
 I 
 
 
 
 ference in volatility. 
 
68 RADIOACTIVE TRANSFORMATIONS 
 
 The family of products of thorium is graphically represented 
 in Fig. 18. 
 
 RADIOTHORIUM 
 
 There has been a considerable difference of opinion as to 
 whether thorium is a true radioactive element or not, i. e., as 
 to whether the activity of thorium is due to thorium itself, or to 
 some active substance normally always associated with it. Some 
 experimenters state that by special methods they have obtained 
 an almost inactive substance giving the chemical tests of tho- 
 rium. Some recent work of Hahn J is of especial importance 
 in this connection. 
 
 THORIUM. rnoK.x. E/viflNariON. THOR.A. THOR.B. 
 
 FIG. 18. 
 Family of thorium products. 
 
 Working with the Ceylon mineral, thorianite, which consists 
 mainly of thorium and 12 per cent of uranium, Hahn was able 
 by special chemical methods to separate a small amount of a 
 substance comparable in activity with radium. This substance, 
 which has been named " radiothorium, " gave off the thorium 
 emanation to such an intense degree that the presence of the 
 emanation could be easily seen by the luminosity produced on a 
 zinc sulphide screen. Thorium X could be separated from it in 
 the same way as from thorium, while the excited activity pro- 
 duced by the emanation decayed with the period of 11 hours 
 characteristic for thorium. The activity of radiothorium seems 
 to be fairly permanent, and it seems probable that this active 
 
 i Hahn : Proc. Roy. Soc., March 16, 1905 ; Jahrbuch. d. Radioaktivitat, II, Heft 3. 
 1905. 
 
 
RADIOACTIVE CHANGES IN THORIUM 69 
 
 substance is in reality a lineal product of thorium intermediate 
 between thorium and thorium X. The radiothorium produces 
 thorium X, which in turn produces the emanation. It still 
 remains to be shown that this active substance can be separated 
 from ordinary thorium, but there can be little doubt that radio- 
 thorium is either the active constituent mixed with thorium, or, 
 what is more probable, that it is a product of thorium. We 
 shall see later that actinium itself is inactive, although it gives 
 rise to a succession of active products remarkably similar in 
 many respects to the family of products observed in thorium. 
 The results of Hahn suggest that the transformation of thorium 
 itself may be rayless, but that the succeeding product, radio- 
 thorium, gives out rays. Further results are required before 
 such a conclusion can be considered as definitely established, but 
 the results so far obtained by Hahn are of the greatest interest 
 and importance. 
 
CHAPTER III 
 THE RADIUM EMANATION 
 
 SHORTLY after the writer l had shown that thorium compounds 
 continuously emit a radioactive emanation, Dorn 2 found that 
 radium possesses a similar property. Very little emanation 
 is emitted from radium compounds in a solid state, but it 
 escapes freely when the radium is dissolved or heated. While 
 the emanations of thorium and radium possess very analogous 
 properties, they can readily be distinguished from each other 
 by the difference in the rates of decay of their activities. While 
 the activity of the thorium emanation decreases to half value 
 in 54 seconds, and practically disappears in the course of 10 
 minutes, that of the radium emanation is far more persistent, 
 for it takes nearly 4 days to be reduced to half value, and is 
 still appreciable after a month's interval. 
 
 In physical and chemical properties the radium emanation 
 is very similar to that of thorium, but, on account of its great 
 activity and comparatively slow rate of change, it has been 
 .studied in more detail than the latter. It has been found possi- 
 ble to isolate it chemically and to measure its volume, as well 
 as to observe its spectrum. The activity and concomitant heat- 
 ing effect, which are enormous in comparison with the amount of 
 matter involved, have drawn strong attention to this substance, 
 for the effects produced are of a magnitude that can neither be 
 easily explained nor explained away. For these reasons, we shall 
 consider in some detail the more important chemical and physical 
 properties of the radium emanation, and the connection that 
 exists between them. The study of this substance will throw 
 additional light on the general theory of radioactivity which has 
 already been developed in the last chapter. 
 
 1 Rutherford: Phil. Mag., Jan., Feb., 1900. 
 
 2 Dorn: Naturforsch. Ges. fur Halle a. S., 1900. 
 
THE RADIUM EMANATION 71 
 
 The salts of radium, generally employed in experimental 
 work, are the bromide and the chloride. Both of these com- 
 pounds emit very little emanation into a dry atmosphere. The 
 emanation produced is stored up or occluded in the mass of the 
 substance, but is released by heating or dissolving the compound. 
 The enormous activity of the emanation set free from radium is 
 very well illustrated by the following simple experiment. 
 
 A minute crystal of the bromide or chloride is dropped into a 
 small wash bottle. A few cubic centimeters of water are added 
 to dissolve the compound, and the bottle is immediately closed. 
 A slow current of air is then sent through the solution, and is 
 carried along a narrow glass tube into the interior of an elec- 
 troscope. If the electroscope is initially charged the leaves are 
 observed to collapse almost immediately after the air reaches it. 
 It is then found impossible to cause a divergence of the leaves for 
 more than a moment. If the emanation is all blown out from 
 the electroscope by a current of air, the leaves are still observed 
 to collapse rapidly, although the emanation has been completely 
 removed. 
 
 This residual activity is due to an active deposit left on the 
 sides of the vessel. In this respect, the emanation of radium 
 possesses a similar property to that of thorium. The activity, 
 however, diminishes more rapidly than in the case of thorium, 
 for most of the electrical effect due to it disappears in a few 
 hours, while in the case of thorium the effect lasts for several 
 days. 
 
 Measurements of the rate of decay of the activity of the ema- 
 nation have been made by several observers. Rutherford and 
 Soddy * stored a quantity of air mixed with emanation in a small 
 gasometer over mercury, and a definite volume was withdrawn 
 at intervals and discharged into a testing vessel such as is shown 
 in Fig. 10. The activity observed in the vessel increased for 
 several hours after the introduction of the emanation on account 
 of the formation of the active deposit. By determining the 
 saturation current immediately after the passage of the emana- 
 tion into the testing cylinder, the quantity of emanation initially 
 
 1 Rutherford and Soddy: Phil. Mag., April, 1903. 
 
72 RADIOACTIVE TRANSFORMATIONS 
 
 present was measured. In this way it was found that the amount 
 of emanation present decreased according to an exponential law, 
 falling to half value in 3.77 days. P. Curie 1 determined the 
 constant of decay of the emanation in a somewhat different way. 
 A large quantity of emanation was introduced into a glass tube, 
 which was then sealed off, and the ionization due to the issuing 
 rays was measured at intervals by an electrometer in a suitable 
 testing vessel. Now it will be seen later that the emanation gives 
 out only a rays, which are completely stopped by a thickness of 
 glass less than ^ mm. ; consequently the rays from the emana- 
 tion were absorbed in the walls of the glass tube. The electrical 
 effect produced in the testing vessel was due entirely to the ft 
 and 7 rays which are emitted from the active deposit produced 
 on the inside of the tube by the emanation. Since after about 
 3 hours the active deposit is in radioactive equilibrium with 
 the emanation, and then decays at the same rate as the parent 
 substance, the intensity of the /3 and 7 rays will diminish at the 
 same rate and according to the same law as the emanation itself. 
 In this way the activity was found to diminish according to an 
 exponential law, falling to half value in 3.99 days. The agree- 
 ment of these periods of decay obtained by different methods 
 shows that the amount of the active deposit is always propor- 
 tional to the amount of emanation present at any time during the 
 life of the emanation. This is one of the proofs that the active 
 deposit is a product of the decomposition of the emanation. 
 
 Further experiments to determine the constant of decay of 
 the emanation have been made by Bumstead and Wheeler, 2 and 
 Sackur. 3 The former found that the activity decreased to half 
 value in 3.88 days, and the latter found the period to be 3.8.6 
 days. We may thus 'conclude that the emanation decays ex- 
 ponentially with a period of about 3.8 days. 
 
 The emanation from radium is almost entirely released 
 boiling a solution of the compound or by aspirating air througl 
 it. The active deposit is left behind with the radium, but thii 
 
 1 P. Curie: Comptes rendus, cxxxv, p. 857 (1902). ' 
 
 2 Bumstead and Wheeler : Amer. Jour. Science, Feb., 1904. 
 
 8 Sackur: Ber. d. d. chem. Ges., xxxviii, No. 7, p. 1754 (1905). 
 
THE RADIUM EMANATION 
 
 73 
 
 disappears after several hours. If the radium solution is then 
 evaporated to dryness, the activity measured by the a rays is 
 found to have reached a minimum of about 25 per cent of the 
 normal value. If kept in a dry atmosphere, the emanation pro- 
 duced from the radium is occluded in its mass, and the activity 
 of the radium consequently increases, reaching its normal steady 
 value after about one month. The recovery curve of the activ- 
 ity of radium from the 25 per cent minimum is shown in Fig. 19. 
 The decay curve of the emanation is added for comparison. 
 
 /oo 
 
 FIG. 19. 
 
 Decay curve of the radium emanation and recovery curve of the activity of radium 
 measured by the a rays, from the 25 per cent minimum. 
 
 As in the case of thorium, the decay and recovery curves 
 are complementary to each other. The activity of the emana- 
 tion falls to half value in about 3.8 days, while half of the lost 
 activity of the radium is recovered in the same interval. 
 
 The activity of the emanation released from the radium is 
 thus given at any time by the equation 
 
74 RADIOACTIVE TRANSFORMATIONS 
 
 while the equation of the recovery curve from the minimum is 
 
 h - I _ 
 ' 
 
 i. e., the amount of the emanation N stored up in the radium 
 after standing for a time t is given by 
 
 where N is the maximum amount. These curves are explained 
 in exactly the same way as the similar curves for thorium. The 
 emanation is an unstable substance which is half transformed 
 in 3.8 days. It is produced at a constant rate by the radium, 
 and the activity of the radium reaches a steady value when the 
 rate of production of fresh emanation balances the rate of dis- 
 appearance of that already formed. 
 
 A radium compound initially freed from emanation will have 
 grown a maximum supply again about one month later, and this 
 process of removal and fresh growth may be continued indefi- 
 nitely. If NO be the number of atoms of emanation presenl 
 ^vvhen in equilibrium, the rate q of supply of fresh atoms oi 
 emanation by the radium is equal to the number lost by its owi 
 decomposition, i. e., 
 
 The value of X thus has a definite physical meaning, for i1 
 represents the fraction of the equilibrium amount of emanatioi 
 supplied per second, as well as the fraction of the emanatioi 
 which breaks up per second. Taking the period of the emana- 
 tion as 3.8 days, the value of X, with the second as the unit oi 
 time is 1/474000, or, in other words, the rate of supply of the 
 emanation per second is 1/474000 of the equilibrium amounl 
 
 This result is well illustrated by a very simple experimenl 
 described by Rutherford and Soddy. A small quantity oi 
 radium chloride in radioactive equilibrium was dropped in I 
 hot water. The accumulated emanation released by solutioi 
 
THE RADIUM EMANATION 75 
 
 was swept with a current of air into a suitable testing vessel, 
 and the saturation current immediately measured. The current 
 so determined is a comparative measure of JVi, the equilibrium 
 amount of emanation stored up in the radium. 
 
 The radium solution was then aspirated with air for some 
 time, to remove the last trace of accumulated emanation, and 
 then allowed to stand undisturbed for 105 minutes. The 
 emanation accumulated in this interval was then swept into a 
 similar testing vessel and the saturation current again deter- 
 mined. This current is a measure of the amount N t of the 
 
 N 
 
 emanation formed in the interval. The ratio was found 
 
 to be .0131, and disregarding the small decay of the emanation 
 during such a short interval, 
 
 N t = q X 105 X 60. 
 It follows that -j=r = 1/480000. 
 
 Allowing for the small decay during the interval, 
 jf = 1/477000. 
 
 From the constant of decay of the emanation we have seen 
 that 
 
 A = J^ = 1/474000. 
 
 The agreement between theory and experiment is thus re- 
 markably close, and is a direct proof that the production of 
 emanation in a solid compound proceeds at the same rate as in 
 the solution. In the former case it is occluded, and in the 
 latter, part is retained in the solution and the rest in the air 
 space above it. 
 
 It is surprising how tenaciously the emanation is held by dry 
 radium compounds. Experiment showed that the emanating 
 power in the solid state was less than one half per cent of the 
 emanating power in solution. Since a radium compound stores 
 
76 RADIOACTIVE TRANSFORMATIONS 
 
 up nearly 500,000 times as much emanation as is produced per 
 second, the result shows that the amount of emanation escaping 
 per second is less than one hundred millionth part of that oc- 
 cluded in the compound. The rate of escape of emanation is 
 much increased in a moist atmosphere and by rise of temperature. 
 
 The recovery curve of a solid radium compound freed from 
 emanation is altered if the conditions allow much of the recov- 
 ered emanation to escape. Under such conditions, the maximum 
 activity is reached more quickly, and is far smaller than the 
 normal activity of a non-emanating compound. 
 
 This property of radium of retaining its emanation is difficult 
 to explain satisfactorily unless it is assumed that there is some 
 slight chemical combination between the emanation and the 
 radium producing it. Godlewski l has suggested that the eman- 
 ation is in a state of solid solution with the parent matter. 
 This point of view is supported by certain observations made by 
 him on the rapidity of diffusion of the product uranium X into 
 a uranium compound. A discussion of his results will be given 
 later in Chapter VII. 
 
 CONDENSATION OF THE EMANATION 
 
 For several years after the discovery of the emanations from 
 thorium and radium, there existed considerable difference of 
 opinion as to their real nature. Some physicists suggested 
 that they were not material, but consisted of centres of force 
 attached to the molecules of gas with which the emanation was 
 mixed, and moving with them. Others held that the emana- 
 tion was a gas present in such minute amount that it was diffi- 
 cult to detect by means of the spectroscope or by direct chemical 
 methods. The objections urged against the material character 
 of the emanation were to a large extent removed by the dis- 
 covery, made by Rutherford and Soddy, 2 that the emanations 
 of thorium and radium possessed a characteristic property of 
 gases inasmuch as they could be condensed from the inactive 
 gas with which they were mixed by the action of extreme cold. 
 
 1 Godlewski: Phil. Mag., July, 1905. 
 
 2 Rutherford and Soddy: Phil. Mag., May, 1903. 
 
THE RADIUM EMANATION 
 
 77 
 
 As a result of a careful series of experiments, it was found that 
 the emanation from radium condensed at a temperature of 
 150 C. The condensation and volatilization points were very 
 sharply defined, and did not differ by more than 1 C. The 
 thorium emanation commenced to condense at about 120 C., 
 but the condensation was not usually completed until a tempera- 
 ture of 150 C. was reached. The probable cause of this in- 
 teresting difference in behavior of the two emanations will be 
 discussed presently. 
 
 If a large amount of emanation is available, the condensation 
 of the radium emanation can readily be followed by the eye. 
 The experimental arrangement is clearly shown in Fig. 20. 
 
 The emanation mixed 
 with air is stored in a 
 small gasometer, and 
 is then slowly passed 
 through a U tube im- 
 mersed in liquid air. 
 This U tube is filled 
 with fragments of wil- 
 lemite, or crystals of 
 barium platinocyanide, 
 which become luminous 
 under the influence of 
 the rays from the ema- 
 nation. If the current of air mixed with emanation is passed 
 very slowly through the tube, the fragments of willemite begin 
 to glow brightly just below the level of the liquid air, and the 
 luminosity can be concentrated over a short length of the tube. 
 This shows that the emanation has been condensed at the tem- 
 perature of liquid air, and is deposited on the walls of the tube 
 and on the surface of the willemite. If the U tube is then 
 partially exhausted and closed with stopcocks, the emanation 
 still remains concentrated for some minutes on the tube and 
 willemite, although the liquid air is removed. When, however, 
 the temperature of the tube rises to 150 C., the emanation is 
 rapidly volatilized, and distributed throughout the tube. This 
 
 RADIUM EMANATION 
 
 FIG. 20. 
 
78 RADIOACTIVE TRANSFORMATIONS 
 
 is observed by the sudden distribution of the luminosity through- 
 out the whole mass of willemite in the U tube. The point of 
 condensation remains brighter than the rest of the tube for some 
 time. This is due to the fact that the emanation, even in the 
 condensed state, has produced the active deposit. When the 
 emanation is volatilized, the active deposit remains behind, and 
 the rays from it cause a greater luminosity at that point. After 
 an hour's interval this difference of luminosity has almost dis- 
 appeared, and the willemite glows throughout with a uniform 
 light. The luminosity can at any time be concentrated at any 
 point by local cooling with liquid air. 
 
 If the U tube is filled with different layers of phosphorescent 
 materials, like willemite, kunzite, zinc sulphide, and barium 
 platinocyanide, the emanation after volatilization is equally 
 distributed, and each layer of material glows with its own pecu 
 liar light. The greenish luminosity of the willemite and barium 
 platinocyanide is not easily distinguishable, except for a differ 
 ence of intensity. The kunzite glows with a deep red color 
 while the zinc sulphide emits a yellow light. There are severa 
 interesting points of distinction between the action of the ray 
 of the emanation and of the active deposit on these substances 
 Unlike the other substances mentioned, the luminosity of zinc 
 sulphide largely disappears at the temperature of liquid air, bu 
 revives at a higher temperature. The a rays produce a markec 
 luminosity in willemite, the platinocyanides, and zinc sulphide 
 but have little or no effect in lighting up kunzite. The latter i 
 sensitive only to the /3 and 7 rays emitted from the active deposit 
 In consequence of this, the kunzite is very feebly luminous when 
 the emanation is first introduced. The light, however, increases 
 in intensity as the active deposit is produced by the emanation 
 and reaches a maximum about three hours after the introduction 
 of the emanation. After barium platinocyanide has been exposec 
 for some time to the action of a large amount of the emanation 
 the crystals change to a reddish tinge, and the luminosity is 
 much reduced. This has been shown to be due to a permanen 
 change in the crystals by the action of the rays. By re-solution 
 and crystallization, the luminosity again returns. 
 
THE RADIUM EMANATION 
 
 79 
 
 Curie and Debierne early showed that glass becomes luminous 
 under the action of the rays from the emanation. This effect 
 is most marked in Thuringian glass, but as a rule the luminosity 
 is feeble compared with that produced in willemite or zinc sul- 
 phide. The glass becomes colored under the action of the rays, 
 and with strong emanation is rapidly blackened. 
 
 The sharpness of the temperature of volatilization of the 
 radium emanation was very clearly illustrated by some experi- 
 ments made by Rutherford and Soddy, using the electric method. 
 
 FIG. 21. 
 
 Determination of the temperature of condensation of the radium emanation 
 by the electric method. 
 
 The emanation collected in the gasometer, B, was condensed in 
 a long spiral copper tube, S, (see Fig. 21) immersed in liquid 
 air, and a slow steady stream of air after passing through the 
 tube entered a small testing vessel, T. After condensation the 
 copper spiral was removed from the liquid air and allowed to 
 heat up very slowly. The temperature was deduced from 
 measurements of the resistance of the copper spiral. Just be- 
 fore the point of volatilization was reached, very little effect 
 was observed in the testing vessel. Suddenly a rapidly increas- 
 ing movement of the electrometer needle was noted, and by 
 
80 RADIOACTIVE TRANSFORMATIONS 
 
 using a large quantity of emanation the rate of movement in- 
 creased in a few moments from several divisions to several 
 hundred divisions per second. The rise of temperature ob- 
 served between the point at which there was practically no 
 escape of the emanation and the point of rapid escape was not 
 more than a fraction of a degree in many cases. 
 
 It has been already pointed out that the temperature of com- 
 plete condensation of the thorium emanation is not at all sharp, 
 but that the condensation in most cases continues over a range of 
 about 30 C. This striking difference in the behavior of the two 
 emanations is, in all probability, due to the small amount of the 
 thorium emanation present in the experiments. The emanation 
 of thorium breaks up at about six thousand times the rate of 
 that of radium. For an equal expulsion of a particles by the two 
 emanations, i. e. for approximately equal electrical effects, the 
 latter must therefore be present in at least six thousand times the 
 amount of the former. In addition, in most of the experiments 
 with the radium emanation, the quantity of emanation was suffi- 
 cient to produce several hundred times the electrical effect ob- 
 served with the small quantity of the emanation obtained from 
 thorium compounds. Thus, in some of the experiments, the 
 quantity of radium emanation present was at least ten thousand 
 times and in many cases more than a million times the 
 amount of the thorium emanation. In fact, it can readily be cal- 
 culated that in the actual experiments not more than 100 atoms 
 of thorium emanation could have been present per cubic centi- 
 metre of gas carried through the copper spiral. Under such 
 conditions, it is not so much a matter of surprise that the 
 emanation of thorium does not show a sharp condensation point, 
 as that the emanation can be condensed at all when so sparsely 
 distributed throughout a volume of gas. 
 
 Diminution of the pressure of the air in the spiral, or the sub- 
 stitution of hydrogen for oxygen as the carrying medium, both 
 tended to cause more rapid condensation. Such an effect is to 
 be expected on the above view, since the rapidity of diffusion 
 ^f the atoms of emanation through the gas is thereby hastened. 
 
 If the thorium emanation should ever be obtained in large 
 
THE RADIUM EMANATION 81 
 
 quantity, there can be little doubt that it will also exhibit com- 
 paratively sharp points of condensation and volatilization. The 
 fact that the thorium emanation begins to condense at a higher 
 temperature (120 C.) than the radium emanation (150 C.) 
 shows that the emanations consist of different types of matter. 
 The emanation of actinium, like the emanations from radium and 
 thorium, may be condensed by passing it through a spiral im- 
 mersed in liquid air, but the rapidity of the decay of its activity 
 (half value in 3.9 seconds) makes an accurate determination of 
 its condensation temperature by the electric method very diffi- 
 cult, since the emanation would lose the greater part of its activ- 
 ity before the stream of gas carrying the emanation could be 
 reduced to the temperature of the spiral. The ease with which 
 the radium emanation is condensed by liquid air has proved of 
 great importance in many recent researches on the emanation. 
 By the use of this property, it has been freed from the gases mixed 
 with it, isolated in a pure state, and its spectrum determined. 
 
 RATE OP DIFFUSION OF THE EMANATION 
 
 If the emanation is introduced at one end of a tube kept at 
 constant temperature, after the lapse of several hours it is found 
 to be distributed in equal amount throughout the volume of the 
 tube. This result shows that the emanation diffuses through 
 the air like an ordinary gas. It has not yet been found possible 
 to determine by a direct method the density of the emanation, 
 as the quantity released from even one gram of pure radium 
 bromide would be too small to be weighed accurately. By com- 
 paring the rate of diffusion of the emanation with that of a 
 known gas, we can, however, obtain a rough estimate of its 
 molecular weight. The rates of interdiffusion of various gases 
 have long been known to decrease with the molecular weight 
 of the diffusing gas. If therefore, for example, we find that the 
 coefficient of interdiffusion of the emanation into air lies between 
 the corresponding values obtained for two known gases, A and 
 B, it is probable that the molecular weight of the emanation is 
 intermediate in value between that of A and B. 
 
 Shortly after the discovery of the radium emanation, Ruther- 
 
 6 
 
82 RADIOACTIVE TRANSFORMATIONS 
 
 ford ancMNIiss Brooks 1 determined its coefficient of interdiffu- 
 sion K into air, and found values lying between K = .07 and 
 K= .09. The method adopted was to divide a long cylinder 
 into equal parts by a movable slide. The emanation was first 
 introduced into one half of the tube, and thoroughly mixed 
 with the air. When temperature conditions were steady, the 
 slide was opened, and the emanation gradually diffused into the 
 other half. The amount % of emanation present in each half of 
 the tube, at any time after opening the slide, was determined 
 by the electric method, and from these data the coefficient of 
 interdiffusion can be calculated. The coefficient of interdiffu- 
 sion of carbon dioxide (molecular weight 44) into air was long 
 ago found to be .142. The emanation thus diffuses into air 
 more slowly than does carbon dioxide into air. For alcohol 
 vapor (molecular weight 77), the value of .2" =.077. Taking 
 the lower value, K= .07, as the more probable value for the 
 radium emanation, it follows that the emanation has a molecu- 
 lar weight greater than 77. 
 
 A number of determinations have since been made, by differ- 
 ent methods, to form an estimate of the molecular weight of the 
 emanation. 
 
 Bumstead and Wheeler 2 measured directly the comparative 
 rates of diffusion of the. emanation and of carbon dioxide through 
 a porous pot. Assuming Graham's law, viz., that the coeffi- 
 cient of interdiffusion is inversely proportional to the square 
 root of the molecular weight, they deduced that the molecular 
 weight of the emanation was about 172. 
 
 Makower, 3 using a similar method, compared the rates of 
 diffusion of the radium emanation through a porous pot with the 
 rates for the gases oxygen, carbon dioxide, and sulphur dioxide, 
 and finally concluded that the emanation had a molecular weight 
 in the neighborhood of 100. Curie and Danne 4 determined the 
 
 1 Rutherford and Miss Brooks : Trans. Roy. Soc., Canada, 1901 ; Chemical News, 
 1902, 
 
 2 Bumstead and Wheeler : Amer. Journ. Sci., Feb., 1904. 
 Makower: Phil. Mag., Jan., 1905. 
 
 * Curie and Danne : Comptes rendus, cxxxvi, p. 1314 (1904). 
 
THE RADIUM EMANATION 83 
 
 rate of diffusion of the emanation through capillary tubes, and 
 obtained a value K = .09, a value somewhat higher than that 
 obtained by Miss Brooks and the writer. 
 
 It is thus seen that all the experiments on diffusion bear 
 out the conclusion that the emanation is a heavy gas with a 
 molecular weight probably not less than 100. It is doubt- 
 ful, however, whether much reliance can be placed on the 
 actual value of the molecular weight deduced in this way, 
 because the emanation exists in minute amount in the gas 
 in which it diffuses, and its coefficient of interdiffusion is 
 compared with that of gases existing in large quantity. The 
 coefficients of interdiffusion may not in such a case be directly 
 comparable. In addition, the rate of diffusion of the eman- 
 ation, which has the properties of a monatomic gas, is coin- 
 pared with the rates of diffusion of gases which have complex 
 molecules. 
 
 If the emanation is considered to be a direct product of radium, 
 and to consist of the radium atom minus one or two a particles, 
 the molecular weight should be not much less than the atomic 
 weight of radium, viz., 225. It is doubtful whether the value 
 of the molecular weight of the emanation can be determined 
 with any certainty until the emanation has been obtained in 
 sufficient amount to determine its density. 
 
 The coefficient of interdiffusion of the thorium emanation 
 into air has been determined by the writer to be about .09. 
 This would suggest that the thorium emanation has a somewhat 
 smaller molecular weight than that of radium. 
 
 The emanation obeys he laws of gases, not only as regards 
 diffusion, but also in other particulars. For example, the 
 emanation divides itself between two connected reservoirs in 
 proportion to their volumes. P. Curie and Danne showed 
 that if one of the reservoirs was kept at a temperature of 
 10 C. and the other at 350 C., the emanation is distributed 
 between them in the same proportion as a gas under the same 
 conditions. 
 
 The emanation thus possesses the characteristic properties of 
 gases, namely, condensation and diffusion. It also obeys at low 
 
84 RADIOACTIVE TRANSFORMATIONS 
 
 temperatures Charles's law, and, as will be seen later, Boyle's 
 law. 
 
 We may thus conclude with confidence that the emanation, 
 while it exists, is a radioactive gas of heavy molecular weight. 
 
 PHYSICAL AND CHEMICAL PROPERTIES OF THE EMANATION 
 
 A number of experiments have been made to determine 
 whether the emanation possesses any definite chemical proper- 
 ties which would enable us to compare it with any other known 
 gas, but so far no evidence has been obtained that the emana- 
 tion is able to combine with other substances. In such experi- 
 ments, the electric method offers a simple and accurate method 
 of determining whether the quantity of the emanation is re- 
 duced under various conditions. In fact, it serves as a rapid 
 and exact method of quantitative analysis of the minute amount 
 of emanation under experiment. 
 
 Rutherford and Soddy l showed that the emanation was not 
 diminished in quantity after condensation by liquid air, or by 
 passage through a platinum tube kept at a white heat by an 
 electric current. A number of experiments were also made in 
 which the emanation was made to pass over a number of re- 
 agents, the emanation being always mixed with, a gas unaffected 
 by the particular reagent. They concluded from these experi- 
 ments that no gas could have survived in unaltered amount the 
 severe treatment to which it had been exposed, except an inert 
 gas of the helium-argon family. 
 
 Ramsay and Soddy 2 found that the quantity of emanation 
 was unchanged after sparking for several hours with oxygen 
 over alkali. The oxygen was then removed by ignited phos- 
 phorus, and no visible residue was left. Another gas was then 
 introduced, and the emanation after mixture with it was with- 
 drawn. Its activity was practically unaltered. A similar re- 
 sult was observed when the emanation was introduced into a 
 magnesium lime tube which was heated for three hours to 
 a red heat. 
 
 1 Rutherford and Soddy: Phil. Mag., Nov., 1902. 
 
 2 Ramsay and Soddy : Proc. Roy. Soc., Ixxii, p. 204 (1903). 
 
THE RADIUM EMANATION 85 
 
 We may thus conclude that the radium emanation in respect 
 to the absence of definite combining properties is allied to the 
 recently discovered inert gases of the atmosphere. On the dis- 
 integration theory, the emanation is supposed to be transformed 
 with the accompaniment of the expulsion of a particles. It is 
 of great importance to settle whether the rate of disintegration 
 is affected by temperature. Any change in the rate of trans- 
 formation would result in a change in the period of decay of 
 the emanation. This point has been examined by P. Curie, 
 who found that the decay of activity was unaffected by con- 
 Atinued exposure of the emanation to temperatures varying be- 
 /tween -180 C. and 450 C. 
 
 This result shows that the transformation of the emanation 
 cannot be considered to be a type of ordinary chemical dissocia- 
 tion, for no reaction is known in chemistry which is independ- 
 ent of temperature over such a wide range. In addition, the 
 transformation of the emanation is accompanied by the expul- 
 sion of a portion of its mass at enormous speed a result never 
 observed in chemical reactions. Such a result suggests that 
 the change that occurs is not molecular but atomic. This view 
 is strongly confirmed by the enormous release of energy during 
 the disintegration of the emanation which will be considered 
 later. 
 
 VOLUME OP THE EMANATION 
 
 It has been seen that the amount of emanation to be obtained 
 from a given quantity of radium reaches a maximum value 
 when the rate of supply of fresh emanation balances the rate of 
 transformation of that already produced. Since this maximum 
 amount of emanation is always proportional to the quantity of 
 radium present, the volume of emanation released from one 
 gram of radium in radioactive equilibrium should have a definite 
 constant value. It was early recognized that the volume of the 
 emanation to be obtained from one gram of radium was very 
 small, but not too minute to be measured. From the data avail- 
 able at the time, the writer l in 1903 calculated that the volume 
 
 1 Rutherford : Nature, Aug. 20, 1903 ; Phil. Mag., Aug., 1905. 
 
86 RADIOACTIVE TRANSFORMATIONS 
 
 of the emanation derived from one gram of radium probably lay 
 between .06 and .6 cubic millimetres at atmospheric pressure 
 and temperature. 
 
 A more accurate deduction can be made from the more recent 
 experimental data of the number of a particles expelled from 
 one gram of radium per second. This number has been deter- 
 mined by the writer by measuring the positive charge communi- 
 cated to a body on which the a rays impinged. Assuming that 
 each a particle carries an ionic charge of 3.4 x 10~ 10 electro- 
 static units, it was deduced that one gram of radium at its mini- 
 mum activity (i. e., when the emanation and its disintegration 
 products were removed) emitted 6.2 x 10 10 a particles per second. 
 If we suppose, as is probably the case, that each radium atom 
 in breaking up gives rise to one atom of the emanation, the 
 number of atoms of emanation produced per second is equal to 
 the number of a particles expelled per second. 
 
 But NO, the maximum number of atoms of emanation stored 
 
 up in radium in radioactive equilibrium, is given by N Q = |> 
 
 A 
 
 where q is the rate of production and X is the decay constant. 
 
 Consequently, the value of No for one gram of radium is 
 6.2 x ID" x 474,000, or 2.94 x 10 16 . 
 
 Now from experimental data it is known that one cubic cen- 
 timetre of any gas at atmospheric pressure and temperature 
 contains 3.6 x 10 19 molecules. Assuming that the molecule 
 of the emanation consists of one atom, the volume of emanation 
 from one gram of radium is 
 
 9 Q9 v 1 16 
 
 36x10" = .0008 c.c., or 0.8 c.mms. 
 
 We shall now consider the changes that may be expected to 
 occur in a volume of pure emanation from the point of view of 
 the disintegration theory. The emanation emits a particles 
 and is transformed into the active deposit, which behaves as 
 a type of non-gaseous matter and attaches itself to the walls 
 of the containing vessel. The amount of emanation decreases 
 exponentially, falling to half value in 3.8 days. We should 
 
THE RADIUM EMANATION 
 
 87 
 
 thus expect the volume of the emanation to shrink, and since 
 the activity of the emanation has decayed to a small fraction of 
 its original value after one month, the volume of 
 the emanation after that interval should be very 
 small. The remarkable way in which these theo- 
 retical conclusions have been verified will now be 
 considered. 
 
 Ramsay and Soddy l attacked the difficult prob- 
 lem of isolating the emanation and determining 
 its volume in the following way. The emanation 
 from 60 milligrams of radium bromide in solu- 
 tion was collected for 8 days, 
 and then drawn off through 
 the inverted siphon, E (see 
 Fig. 22), into the explosion 
 burette, F. The radium in 
 the solution produces hydro- 
 gen and oxygen at a rapid 
 rate, and the emanation was 
 initially removed with these 
 gases. After explosion, the 
 slight excess of hydrogen 
 mixed with the emanation was 
 left for some time in contact 
 with caustic soda, placed in 
 the upper part of the burette, 
 in order to remove the car- 
 bon dioxide present. In the 
 meantime, the upper part of 
 the apparatus had been ex- 
 hausted as completely as possi- 
 ble. The connection with the 
 mercury pump was then closed, 
 and the hydrogen and ema- 
 nation allowed to enter the apparatus, passing over a phos- 
 phorus pentoxide tube, D, to remove all trace of water vapor. 
 
 1 Kamsay and Soddy : Proc. Koy. Soc., Ixxiii., p. 346 (1904). 
 
 FIG. 22. 
 
 Apparatus of Kamsay and Soddy for 
 
 determining the volume of the 
 
 radium emanation. 
 
88 
 
 RADIOACTIVE TRANSFORMATIONS 
 
 The emanation was condensed in the lower part of the tube B, 
 which was surrounded by liquid air. The process of condensa- 
 tion of the emanation at B was made evident by the brilliant 
 luminosity of the lower part of the tube. The mercury of the 
 burette was allowed to run to A, and the tube AB again com- 
 pletely pumped out. The connection with the pump was again 
 closed, the liquid was removed, and the volatilized emanation 
 forced into the accurately calibrated capillary tube A. Obser- 
 vations were then made for a space of several weeks on the 
 variation in volume of the emanation. The results are shown 
 in the following table : 
 
 Time. 
 
 Volume. Time. 
 
 Volume. 
 
 Start 
 
 0.124 c.mms. 
 
 7 days 
 
 0.0050 c.mms. 
 
 Iday 
 
 0.027 " " 
 
 9 
 
 0.0041 " 
 
 3 
 
 0.011 " " 
 
 11 
 
 0.0020 " " 
 
 4 
 
 0.0095 " 
 
 12 
 
 0.0011 " 
 
 6 
 
 0.0063 " " 
 
 
 
 The volume decreased, and after four weeks only a minul 
 bubble remained, but this retained its luminosity to the Ias1 
 During this time, the tube was colored a deep purple by th< 
 rays. This caused difficulties in readings of the volume, and 
 strong source of light was found necessary. Ramsay and Soddj 
 consider that the apparent sudden decrease during the first da] 
 may have been due to the mercury sticking in the capillary tube. 
 Taking the readings after one day, the volume of the emanatioi 
 is found to shrink approximately according to an exponent^ 
 law, decreasing to half value in about 4 days. This is aboul 
 the rate of decrease of volume to be expected from theoretic* 
 considerations. Another experiment was made with a fresl 
 supply of emanation, but a very surprising difference was note( 
 The gas had an initial volume of 0.0254 c.mm. at atmospheri< 
 pressure, and a special series of experiments was made to detei 
 mine the volume occupied by the gas in the capillary tube 
 varying pressures. The emanation was found to obey Boyle's 
 law within the limit of experimental error. Unlike its behavic 
 in the first experiment, however, the volume occupied by the 
 
THE RADIUM EMANATION 89 
 
 in the capillary tube, instead of shrinking, steadily increased, 
 and 23 days later was about 10 times the initial value. At the 
 same time, bubbles commenced to appear in the mercury column 
 below the level of the gas. 
 
 Further experiments are necessary in order to elucidate the 
 contradictions observed in these two experiments. It will be 
 seen later that the gas helium is a transformation product of 
 the emanation. This appears to have been absorbed in the 
 walls of the tube in the first experiment. Such a result is not 
 unexpected, for there is considerable evidence that the a parti- 
 cles expelled from the radioactive products consist of helium 
 atoms projected with great velocity. Most of these atoms 
 would be buried in the walls of the glass tube to an average 
 depth of about .02 mm., and their diffusion back into the gas 
 may depend on the kind of glass employed. The most plausi- 
 ble explanation is that the helium after absorption by the walls 
 of the glass capillary diffused back into the gas in the second 
 experiment, but not in the first. 
 
 Ramsay and Soddy concluded from their experiments that 
 the maximum volume of the emanation released from one gram 
 of radium was slightly greater than one cubic millimetre at 
 standard pressure and temperature. 
 
 The theoretical and calculated amounts 0.8 and 1 c. mm., 
 respectively, are thus in very good agreement, and indicate the 
 general correctness of the theory on which the calculations are 
 based. 
 
 SPECTRUM OF THE EMANATION 
 
 After the isolation of the emanation, and determination of its 
 volume, a number of experiments were made by Ramsay and 
 Soddy to determine its spectrum. In some of the experiments 
 several apparently new bright lines were seen for a moment, 
 but these rapidly vanished in consequence of the liberation of 
 hydrogen in the tube. Ramsay and Collie l continued the ex- 
 periments, and were finally successful in obtaining the spectrum 
 of the emanation, which lasted for a sufficient interval to deter- 
 mine rapidly the wave-lengths of the more obvious lines by 
 
 1 Ramsay and Collie: Proc. Roy. Soc., Ixxiii., p. 470 (1904). 
 
90 RADIOACTIVE TRANSFORMATIONS 
 
 means of eye measurements. The spectrum, however, soon 
 faded, and was finally completely masked by that of hydrogen. 
 They state that the spectrum was very brilliant and consisted 
 of a number of bright lines, the spaces between being perfectly 
 dark. The spectrum bore a striking resemblance in general 
 character to the spectra of the inert gases of the argon family. 
 On repeating the experiment with a fresh supply of emanation, 
 many of the bright lines were seen again, while some new lines, 
 not observed in the first spectrum, made their appearance. 
 They conclude that the emanation undoubtedly has a definite 
 and well-marked spectrum of bright lines. 
 
 HEAT EMISSION OF THE EMANATION 
 
 One gram of radium in radioactive equilibrium continuously 
 emits heat at the rate of about 100 gram calories per hour. If 
 the emanation is released from the radium by solution or heat- 
 ing, the heating effect of the radium decreases to a minimum of 
 about 25 per cent of the original value, and then as new ema- 
 nation is formed it gradually increases, reaching its old value 
 after a month's interval. The vessel containing the emanation 
 released from the radium is found to emit heat at a rapid rate, 
 and, three hours after removal, gives out about 75 per cent of 
 the heat emitted by the original radium. The rate of heat 
 emission of the emanation decays at the same rate as it loses 
 its activity, i. e., it falls to half value in about 4 days. Th< 
 curves of decrease of heating effect of the emanation and oi 
 recovery of the heating effect of the radium are, like the activ- 
 ity curves, complementary to each other. The heat emissioi 
 of the two together is always equal to that of the radium 
 radioactive equilibrium. 
 
 The heat emission of the tube containing the emanation 
 not due to the emanation alone, but also to the active deposit 
 formed from the emanation. The laws controlling the heat 
 emission of radium and its products will be considered moi 
 completely in Chapter X. 
 
 It is thus seen that the emanation, together with its trans 
 formation products, is responsible for about three quarters 
 
THE RADIUM EMANATION 91 
 
 the heat emission of radium. It is difficult to disentangle the 
 heating effect of the emanation from that of its rapidly chang- 
 ing products, but there is no doubt that it supplies about one 
 quarter of the total heating effect of the radium. 
 
 Thus one cubic millimetre of the emanation the maximum 
 amount released from one gram of radium itself emits heat at 
 the rate of 25 gram calories per hour. Now the heating effect 
 of the emanation falls off at the same rate as its activity. The 
 total heat emission of the emanation during its life is given 
 by Ql\. The value of X, with the hour as the unit of time, 
 is 1/132, and since Q = 25, the total heat emitted by the emana- 
 tion is 3300 gram calories. If we include with that of the 
 emanation the heating effects of its subsequent products, the 
 total heat emitted from the emanation tube is about three times 
 this amount, or 9900 gram calories. This corresponds to a vol- 
 ume of the emanation of about one cubic millimetre. The total 
 heat released from one cubic centimetre of the emanation and its 
 products is thus about ten million gram calories. 
 
 Now in the union of hydrogen with oxygen to form water 
 more heat is emitted, weight for weight, than in any other known 
 chemical reaction. In the explosion of 1 c.c. of hydrogen with 
 I c.c. of oxygen to form water, 3 gram calories of heat are 
 emitted. We thus see that the transformation of the emanation 
 is accompanied by nearly four million times as much heat as is 
 given out by the union of an equal volume of hydrogen with 
 oxygen to form water. 
 
 If we assume that the atom of the emanation has 200 times 
 the mass of the hydrogen atom, it can readily be calculated 
 that one pound weight of the emanation would emit energy at 
 a rate corresponding to 10,000 horsepower. This evolution of 
 energy would fall off exponentially, but, during the life of the 
 emanation, the total energy released would correspond to about 
 60,000 horsepower-days. 
 
 These figures bring out in a striking way the enormous evolu- 
 tion of heat accompanying the changes in the emanation. The 
 amount is of quite a different order of magnitude from that 
 absorbed or released in the most violent chemical reactions. 
 
92 RADIOACTIVE TRANSFORMATIONS 
 
 We shall see later (Chapter X) that probably every radio- 
 active product which expels a particles emits an amount of 
 heat of the same order of magnitude as that emitted by the 
 emanation. In fact, it will be shown that this evolution of heat 
 is a necessary accompaniment of their radioactivity, for the heat 
 is a measure of the kinetic energy of the a particles expelled 
 from the emanation and its products. 
 
 DISCUSSION OF RESULTS 
 
 We may now briefly summarize the properties of the radium 
 emanation discussed in this chapter. (1) The emanation is a 
 heavy gas which does not combine with any substances, but 
 appears to be allied in general properties with the inert group 
 of gases of which helium and argon are the best known ex- 
 amples. (2) It diffuses like a gas of high molecular weight 
 and obeys Boyle's law. (3) It has a definite spectrum of 
 bright lines analogous to the spectra of the inert gases. (4) It is 
 condensed from a mixture of gases at a temperature of 150 C. 
 (5) Unlike ordinary gases, the emanation is not permanent, but 
 undergoes transformation according to an exponential law. The 
 volume of the emanation consequently decreases at the same 
 rate as it suffers disintegration, i. e., its volume shrinks to half 
 value in 3.8 days. The transformation of the emanation is 
 accompanied by the expulsion of a particles, and results in the 
 appearance of a new series of non-gaseous substances deposited 
 on the surface of bodies. The properties of the active deposit, 
 and the changes occurring in it, will be discussed in detail in 
 the next chapter. 
 
 The emanation, weight for weight, is about one hundred 
 thousand times as active as the radium from which it is derived. 
 On account of its enormous activity, it glows in the dark and 
 causes a brilliant phosphorescence in many substances. The 
 rays quickly color glass, quartz, and other bodies, and produce 
 a rapid evolution of hydrogen and oxygen in a water solution. 
 The transformation of the emanation is accompanied by an 
 enormous evolution of heat, of an order one million times 
 greater than that observed in any chemical reaction. 
 
THE RADIUM EMANATION 93 
 
 We have seen that the emanation and its subsequent prod- 
 ucts are responsible for three quarters of the activity of 
 radium measured by the a rays. The emanation itself does 
 not emit /3 or 7 rays, but these arise from one of its subsequent 
 products. Consequently the & and 7 ray activity of radium is 
 almost completely removed by depriving it of its emanation, 
 provided that several hours have been allowed to elapse in order 
 that the active deposit left behind with the radium may lose 
 its activity. 
 
 The emanation, with its subsequent products, thus contains 
 the concentrated essence of the radioactivity of radium. A tube 
 containing the radium emanation has all the radioactive proper- 
 ties of radium in equilibrium. It emits a, ft, and 7 rays, evolves 
 heat, and produces luminosity in many substances. Radium 
 itself, freed from the emanation and the active deposit, emits 
 only a rays. Its activity and heating effect under such condi- 
 tions is only one quarter of its usual value, when in radioactive 
 equilibrium. 
 
 The emanation is produced from radium at a constant rate, 
 and appears to be a direct disintegration product of the radium 
 atom. Following the same line of argument previously con- 
 sidered, it may be supposed that a minute fraction of the total 
 number of radium atoms explode every second, each violently 
 ejecting an a particle. The radium atom, minus an a particle, 
 becomes the new substance the emanation. The atoms of 
 the emanation are far more unstable than those of radium itself, 
 and break up with the expulsion of a particles at such a rate 
 that half of the particles are transformed in 3.8 days. After 
 the expulsion of an a particle, the emanation turns into the 
 active deposit. 
 
 The transformations, so far considered, and the rays emitted, 
 are graphically illustrated below: 
 
 a particle 
 
 Radium a particle 
 
 EmanatioD 
 
 * Active deposit 
 
94 RADIOACTIVE TRANSFORMATIONS 
 
 The remarkable differences in the chemical and physical 
 properties of a disintegration product and its parent substance 
 are strikingly illustrated by the comparison of radium with its 
 emanation. Radium is a solid substance of atomic weight 225, 
 closely allied in ordinary chemical properties with barium. It 
 has a definite well-marked spectrum analogous in many respects 
 to the spectra of the rare earths. It is non-volatile at ordinary 
 temperatures, and apart from its radioactivity has all the proper- 
 ties of a new element very analogous to barium. On the other 
 hand, the emanation is an inert gas which cannot be made to 
 combine with any substance. Its spectrum of bright lines is 
 similar in general appearance to the spectra of the helium-argon 
 family of gases. It is condensed at a temperature of 150 C. 
 Apart from its radioactivity, the properties of the emanation 
 are thus entirely different from those of the parent radium, and, 
 if we had no proof of its production by radium, there would be 
 no reason to believe they were in any way connected with each 
 other. 
 
CHAPTER IV 
 TRANSFORMATION OF THE ACTIVE DEPOSIT OF RADIUM 
 
 IN the previous chapter attention has been drawn to the fact 
 that all bodies surrounded by the radium emanation become 
 coated with an invisible active deposit, possessing physical and 
 chemical properties which sharply distinguish it from the 
 emanation. This property of radium of " exciting " or " in- 
 ducing " activity in neighboring bodies was first observed by 
 P. Curie, 1 and has in recent years been the subject of a number 
 of investigations. 
 
 In this chapter, the transformations taking place in this 
 active deposit will be discussed, and it will be shown that, in 
 general, the deposit consists of a mixture of three distinct sub- 
 stances called radium A, B, and C. Radium A arises directly 
 from the transformation of the emanation, radium B arises from 
 radium A, and radium C from radium B. 
 
 The three products are thus derived by the successive dis- 
 integration of the emanation. The analysis of these stages is 
 somewhat more difficult than in the case of two changes already 
 considered for thorium, but can be attacked by the same general 
 methods. 
 
 The active deposit of radium is analogous in many respects 
 to the corresponding deposit produced by the thorium emana- 
 tion. It is a material substance, which, in the absence of an 
 electric field, is deposited from the gas on the surface of all 
 bodies in contact with the emanation. In a strong electric field 
 it is mostly concentrated on the negative electrode. In this 
 respect it behaves similarly to the active deposit of thorium. 
 The active matter can be partly removed from a platinum 
 wire -by solution in hydrochloric acid, and remains behind 
 
 1 M. and Mme. Curie, Comptes rendus, cxxix, p. 714 (1899). 
 
96 RADIOACTIVE TRANSFORMATIONS 
 
 on the dish when the acid is driven off by heat. By using 
 the emanation from about 10 milligrams of radium bromide, a 
 wire can be made intensely active. It causes brilliant fluores- 
 cence on a screen of willemite or zinc sulphide brought near it. 
 The deposit is entirely confined to the surface of a conductor. 
 If a strongly active wire is drawn across a screen of willemite 
 or other substance which lights up under the action of the 
 rays, a bright luminous trail is left behind. This is due to the 
 removal of some of the deposit by the particles of the screen 
 over which it has been rubbed. The luminosity left behind 
 gradually decreases, and is very small after 3 hours. The re- 
 moval of the active deposit by rubbing is also easily shown by 
 bringing near an electroscope a piece of cloth which has been 
 drawn over the active wire. The electroscope is discharged 
 almost instantly, and this discharging property persists, but with 
 diminishing amount, for several hours. 
 
 In the case of a short-lived emanation like that of thorium, 
 the excited activity, in the absence of an electric field, is 
 greatest on bodies placed near the emanating thorium com- 
 pound. This result is to be expected, since the emanation is 
 decomposed before it has time to diffuse far from its source. 
 On the other hand, in a similar enclosure containing radium as 
 a source of emanation, the excited activity is produced on all 
 bodies placed in the vessel. In this case the life of the emana- 
 tion is long compared with the time taken for the emanation to 
 be distributed by the processes of diffusion to all parts of the 
 enclosure. 
 
 Bodies which are completely screened from the direct radia- 
 tion of the radium become active. This is clearly brought out 
 in an experiment made by P. Curie, which is shown in Fig. 23. 
 
 A small open vessel, a, contained a radium solution, giving 
 off emanation at a constant rate. This was placed in a closed 
 vessel in which plates A, B, C, D, E were fixed in various 
 positions. After a day's exposure, all the plates on removal 
 were found to be active, even that in the position D, com- 
 pletely shielded from the direct radiation of the radium by a 
 lead block P. 
 
TRANSFORMATION OF RADIUM 
 
 97 
 
 The amount of activity per unit area on a plate in a given 
 position is independent of the material of the plate. A plate 
 of mica becomes just as active as one of metal. The amount 
 of excited activity on a given area depends to some extent on 
 the free space in the neighborhood. The lower surface of the 
 plate A, for example, would be less active than the upper sur- 
 face, since the active deposit on the lower side arises mainly 
 from the small volume of emanation between it and the en- 
 closure, while the upper surface of the plate gains the active 
 deposit generated in a much larger volume. 
 
 The emanation from 
 several milligrams of 
 radium bromide causes 
 so great an activity on 
 a wire or metal plate 
 exposed in its presence, 
 that the ionization cur- 
 rent produced by it can 
 be readily measured by a 
 sensitive galvanometer. 
 With such intensely 
 active plates, a large 
 voltage is required to 
 produce a saturation 
 current through the gas 
 unless the plates of the 
 testing vessel are placed 
 close together. 
 
 We shall first consider the evidence in support of the view 
 that the active deposit is a disintegration product of the 
 radium emanation. If some radium emanation is introduced 
 into a cylindrical testing vessel such as is shown in Fig. 10, 
 and the ends closed, the activity, measured by the saturation 
 current through the gas, increases with the time for several 
 hours, generally reaching about twice the value observed at the 
 moment of introduction of the emanation. The comparative 
 increase, however, varies to some extent with the dimensions 
 
 7 
 
 FIG. 23. 
 
 Distribution of excited activity on bodies in the 
 presence of the radium emanations. 
 
98 RADIOACTIVE TRANSFORMATIONS 
 
 of the testing vessel, on account of the difference in penetrating 
 power of the a rays emitted by the various products. 
 
 When the emanation is blown out, the active deposit is left 
 behind, and loses the greater part of its activity in a few hours. 
 This property of producing an active deposit is not shown by 
 radium which has been freed from emanation, but belongs 
 to the emanation alone. The excited activity produced in 
 bodies is directly proportional to the amount of emanation 
 present, no matter how old the emanation may be. For ex- 
 ample, if the emanation, which still remains after being stored 
 in a gas holder for a month, is passed into a testing vessel, 
 excited activity is still produced, and in an amount which bears 
 the same ratio to the activity of the emanation present as for a 
 new sample of emanation tested immediately after its release 
 from radium. 
 
 The constancy of this ratio between the amount of the ema- 
 nation present and the amount of active deposit produced is 
 at once explained if the emanation is the parent of the active 
 deposit. For example, suppose that a body is exposed to a 
 constant supply of emanation. The activity imparted to the body 
 reaches a steady limit after about 5 hours. There is, then, a state 
 of equilibrium between the active deposit and the emanation. 
 Under such conditions, the number of atoms of radium A which 
 break up per second must equal the number of new atoms 
 of radium A supplied per second by the decomposition of the 
 emanation. This in turn is equal to the number of atoms of 
 emanation which break up per second. A similar result also 
 holds true for radium B and C. Since the number of atoms of 
 any individual product which break up per second is always 
 proportional to the total number present, it is seen that the 
 equilibrium number of atoms of radium A must always be pro- 
 portional to the number of atoms of emanation. If X is the 
 constant of decay of the emanation and X^, XB, X c , the constants 
 for radium A, B, and C, respectively, then the equilibrium 
 amounts N A , N B , NC, respectively, of the three products are 
 given by the equations 
 
 *A&A = *B N B = A c ^ c = X.2V, 
 
TRANSFORMATION OF RADIUM 99 
 
 where N is the total number of atoms of emanation present. 
 When a state of equilibrium has been reached, the number of 
 atoms of each product present will be different, being directly 
 proportional to the period of each product. A rapidly changing 
 substance will consequently be present in less amount than a 
 slowly changing one. 
 
 After introducing the emanation into a closed vessel, its 
 amount, as we have seen, decreases exponentially. Since, how- 
 ever, the periods of the products of the active deposit are small 
 compared with that of the emanation itself, the amount of the 
 active deposit will, after a few hours, nearly reach an equi- 
 librium value, and will then decrease pari passu with the 
 emanation. 
 
 The excited activity will thus fall off at the same rate as the 
 activity of the emanation. This proportion has been utilized, 
 as we have already seen, by Curie and Danne, to determine the 
 constant of decay of the emanation by measurement of the y3 
 and 7 rays which escape from the active deposit through the 
 walls of a closed vessel containing the emanation. 
 
 ACTIVITY CURVES OF THE ACTIVE DEPOSIT 
 
 We shall now consider in detail the variation with the time 
 of the activity of this deposit under different conditions. The 
 experimental results are at first sight very complicated, for the 
 activity curves not only vary remarkably with the time of ex- 
 posure to the emanation, but also depend on whether the a, /?, 
 or 7 rays are used as a means of measurement. It is thus very 
 necessary in each case to specify carefully, not only the time 
 of exposure to the emanation, but also the type of rays used for 
 measurement. 
 
 The decay curves of the active deposit are independent of 
 the nature and size of the body that has been made active and 
 of the amount of emanation to which it has been exposed. If 
 a wire is to be made active the arrangement shown in Fig. 24 
 is very suitable. 
 
 A solution of radium is placed in a vessel closed by a rubber 
 stopper. The emanation collects in the air space above the 
 
100 
 
 RADIOACTIVE TRANSFORMATIONS 
 
 w 
 
 solution. The thin wire, W, to be made active is fixed into 
 a fine hole bored in the end of a central rod. This rod slips 
 freely through an ebonite cork fixed in a brass tube, B. A 
 platinum wire P passes through the rubber cork and dips into 
 the solution. 
 
 The platinum wire is in metallic connection with the brass 
 tube. The central rod is connected with the negative pole of 
 
 a battery of 300 or 400 volts, and 
 the platinum wire with the positive. 
 Under such conditions, the moist 
 walls of the glass vessel, the solu- 
 tion, and the tube B, are charged 
 positively, and the wire W is the 
 only negatively charged body in 
 the presence of the emanation. 
 The active deposit is consequently 
 concentrated upon it, and, in the 
 presence of a large amount of ema- 
 nation, the activity of the wire 
 becomes very great. 
 
 After introducing the wire, a 
 little hard wax is run round the 
 top of the rod, to prevent the es- 
 cape of the emanation. When 
 the wire has been exposed for the 
 FlG - 24 - interval required, the rod is re- 
 
 Arrangement for concentrating the moved and the active wire released. 
 
 active deposit derived from the g ince ^ fine wire ig of smaller 
 radium emanation on a small 
 
 negatively charged wire. diameter than the rod, the wire 
 
 need not touch the side during 
 removal, so that none of the active deposit is rubbed off. 
 
 In order 4o test the variation of the a ray activity of this wire 
 with time, it is attached to the end of a brass rod forming the 
 central electrode of a testing vessel such as is shown in Fig. 10. 
 
 If a greater surface is to be made active, a sheet of metal is 
 placed in a glass tube closed at both ends. The emanation 
 is introduced after first exhausting the vessel, and the active 
 
TRANSFORMATION OF RADIUM 
 
 101 
 
 matter is then deposited upon the metal by the process of dif- 
 fusion. After removal, the activity of the plate is tested elec- 
 trically, using a parallel plate apparatus similar to that described 
 in Fig. 9. 
 
 a RAY CURVES 
 
 We shall first consider the decay of activity, measured by 
 the a rays, for a body exposed for a short time in the presence 
 of the emanation. The time of exposure not more than one 
 
 100 
 
 Decay 
 
 of Excited 
 measured 
 
 activitjy 
 by ex 
 
 of Rad 
 
 um 
 
 rays. 
 
 Sho 
 
 Exposure (i mirute) 
 
 ID 
 
 2O 30 40 5O OO 7O 
 
 Time in Minutes. 
 FIG. 25. 
 
 minute is supposed to be short compared with the period of 
 the changes hi the active matter. The results are shown in 
 Fig. 25, Curve BB, the maximum activity immediately after 
 removal being taken as 100. 
 
 The activity at first decreases very nearly according to an 
 exponential law, falling off to half value in about 3 minutes. 
 After 20 minutes, the activity is less than 10 per cent of the 
 initial value, and remains nearly constant for a further 20 
 minutes, and then gradually decays. After several hours the 
 
102 RADIOACTIVE TRANSFORMATIONS 
 
 activity again decreases nearly exponentially with a period of 
 28 minutes. 
 
 In the same figure (Curve AA) is shown the a ray decay 
 curve for a long exposure. The time of exposure in this case 
 (about 5 hours will suffice) is supposed to be sufficient to allow 
 the active deposit and the emanation to have very nearly 
 reached a stage of radioactive equilibrium. There is initially 
 a rapid decay with a 3 minute period, and then a gradual de- 
 crease at a slower rate than is given by an exponential law. 
 After about 5 hours the decay curve is nearly exponential, fall- 
 ing to half value in about 28 minutes. 
 
 The initial rapid change with a 3 minute period is due to 
 the product radium A. The final exponential decay with a 
 28 minute period shows that another product, having a 28 
 minute period, is also present. Before discussing the explana- 
 tion of the intermediate portion of the two curves the activity 
 curves measured by the /3 and 7 rays will first be considered. 
 
 /3 RAY CURVES 
 
 In order to determine the ray curves, an electroscope was 
 used. The active plate or wire was placed under the base of 
 the electroscope, which was covered with a sheet of aluminium 
 of sufficient thickness to absorb all the a rays. The discharge 
 produced in the electroscope is then due to the /3 and y rays 
 together, the effect of the former preponderating. The curve 
 in Fig. 26 shows the variation of the /? ray activity with time, 
 for a wire which had been exposed for one minute in the 
 presence of a large amount of emanation. It will at once be 
 observed that the curve is entirely different in character from 
 the corresponding a ray curve shown in Fig. 25. The ft ray 
 activity is small at first, but increases with time, reaching a 
 maximum after about 35 minutes. Several hours later it decays 
 nearly exponentially with a period of 28 minutes. 
 
 The /3 ray curve for a long exposure to the emanation is 
 shown in Fig. 27. 
 
 The curve is very different in shape from the short exposure 
 curve. The activity does not increase initially, but falls, first 
 
TRANSFORMATION OF RADIUM 
 
 103 
 
 slowly and then more quickly. Finally, as in the other cases, 
 it decreases exponentially with a period of 28 minutes. 
 
 7 RAY CURVES 
 
 The curves for a short and long exposure measured by the 
 7 rays alone are identical with those obtained for the ft and 
 7 rays together. The measurements were made with an elec- 
 troscope, the rays passing through about 1 cm. of lead before 
 
 100 
 
 8C 
 
 60 
 
 Short 
 
 j Curvj 
 Expof ure. 
 
 40 
 
 20 
 
 O 15 30 45 60 75 90 105 120 
 
 Time in Minutes. 
 
 FIG. 26. 
 
 Variation of the activity, measured by the # rays, of a body exposed for 
 a short interval to the radium emanation. 
 
 entering the electroscope. This insures that the ft as well as 
 the a rays are cut off completely. : 
 
 The identity of the ft and 7 ray curves shows that the two 
 kinds of rays always occur in the same proportion. This rela- 
 tion is a strong argument in favor of the view that the 7 rays 
 are a type of X-rays, which are set up at the moment of the ex- 
 pulsion of the ft particle from radioactive matter. This ratio 
 between the intensities of the two kinds of rays has been shown 
 to hold in every case so far examined, and suggests that the 
 
104 
 
 RADIOACTIVE TRANSFORMATIONS 
 
 7 rays bear the same relation to the (S rays that the X-rays bear 
 to the cathode rays. 
 
 THEORY OF SUCCESSIVE CHANGES IN RADIUM 
 
 We shall show later that the peculiarities of the decay curves 
 of the active deposit of radium for any time of exposure, whether 
 
 100 
 
 80 
 
 60 
 
 Dec 
 
 ay of Excited A 
 
 of 
 
 tadium. 
 
 fcivlfcy 
 
 o 20 40 60 So 100 120 
 
 Time in Minutes. 
 
 FIG. 27. 
 
 Variation of the activity, measured by the or y rays, of a body exposed 
 for a long interval to the radium emanation. 
 
 the activity is measured by the a, /3, or 7 rays, can be satisfac- 
 torily explained on the following assumptions : 
 
 (1) That the emanation is transformed into a product called 
 radium A, which emits only a rays, and has a period of 
 minutes. 
 
 (2) That radium A is transformed into radium B, which has 
 a period of 28 minutes, and is transformed without the emis- 
 sion of a, /3, or 7 rays. In other words, radium B is a rayl< 
 product. 
 
TRANSFORMATION OF RADIUM 105 
 
 (3) Radium B is transformed into radium C, which has a 
 period of 21 minutes, and emits during its transformation a, /:?, 
 and 7 rays. 
 
 We thus have to deal with the problem of three successive 
 changes. Since, however, the first product, radium A, is 
 rapidly transformed with a 3 minute period, the amount of it 
 remaining, for example, 21 minutes after removal, is only 1/128 
 of the initial amount. 
 
 For simplicity, therefore, in the discussion of the activity 
 curves measured by the /3 rays, we shall for the moment disre- 
 gard the first rapid change, and suppose that the emanation is 
 transformed directly into radium B. As a matter of fact, it is 
 found that the experiments agree better with theory if the first 
 transformation is disregarded altogether. A possible explana- 
 tion of this peculiarity in the curves will be considered later. 
 
 In the discussion of the activity curves for the active deposit 
 of thorium, it has been shown that the experimental curve for 
 a short exposure may be satisfactorily explained if the emana- 
 tion is supposed to change into the rayless product, thorium A, 
 which has a period of 11 hours. This in turn is transformed 
 into thorium B, which emits a, /3, and 7 rays, and has a period 
 of about 1 hour. These results deduced from analysis of the 
 activity curves have been completely substantiated by experi- 
 ments in which the products thorium A and B have been sepa- 
 rated from each other by various physical and chemical methods. 
 
 The case of radium is very analogous, for, disregarding the 
 first 3 minute change, the product radium B emits no rays, 
 but changes into radium C, which emits a, /:?, and 7 rays. 
 
 We shall now consider the theory of two successive changes 
 of the character explained above. 
 
 Let Xj, X 2 be the constants of change of the products radium B 
 and C respectively. 
 
 Let P and Q be the number of atoms of B and C respectively 
 present at any time after removal from the emanation. 
 
 Two general cases will first be considered, corresponding to 
 a short and long exposure of a body to the radium emanation. 
 
106 RADIOACTIVE TRANSFORMATIONS 
 
 CASE OF A SHORT EXPOSURE 
 
 The matter initially deposited is supposed to be all of one 
 kind, radium B. Let n be the number of particles of B that 
 have been deposited. The number, P, of these remaining at 
 any time , after removal, is given by 
 
 P = n e-M. 
 
 We have shown on page 50 that the rate of change of the 
 number Q of atoms of C existing at any time, , after removal, 
 is given by 
 
 dQ -\p AO 
 
 -X,f -\ 2 Q (1) 
 
 = X 1 n er-W \tQ. 
 
 The solution of this equation (see page 51) shows that Q is 
 given by 
 
 Q = -* 
 
 The number of atoms of P and Q existing at any time after 
 removal is shown in Fig. 28. The initial number of atoms of 
 B deposited is supposed to be 100. The exponential curve, BB, 
 expresses the amount of B remaining unchanged at any time. 
 The curve CC shows the number of atoms of radium C existing 
 at any time. The periods of the changes of B and C are about 
 28 and 21 minutes respectively, so that 
 
 A! = 4.13 X 10~ 4 (sec.)" 1 , A,, = 5.38 X 10- 4 (sec.)- 1 . 
 
 The amount of radium C, initially zero, increases to a maximum 
 in about 35 minutes, and then diminishes, and about 5 hours later 
 decays exponentially, with a period of 28 minutes. The amount 
 of C will thus decrease, not according to its own period, but 
 according to the longer period of the rayless product. This is 
 easily shown from the equation for , which may be expressed 
 in the form 
 
 After 7 hours, e-(^-W = .043, 
 
TRANSFORMATION OF RADIUM 
 
 107 
 
 and is thus almost negligible. Q then varies very nearly as 
 e~^*, i. e., according to the period of the rayless product. 
 
 Since B does not emit rays and C does, the value of Q at any 
 time is proportional to the activity of the mixture of products 
 B and C. 
 
 too 
 
 BO 
 
 t> 
 
 8 
 
 5- 
 
 1 
 
 20 
 
 20 
 
 /CO 
 
 /SO 
 
 40 00 8O 
 
 T/MC tfi MMUTZ9. 
 
 FIG. 28. 
 
 Theoretical curves showing the variation of the number of atoms of radium B 
 and radium C when the matter present initially consists only of radium B. 
 
 The curve CC should thus be identical in form with the 
 curves for a short exposure measured by the /B or 7 rays, and 
 within the limits of experimental error this is found to be so. 
 
 CASE OF A LONG EXPOSURE 
 
 Suppose that P and Q are the equilibrium numbers of atoms 
 of B and C present after a long exposure to the emanation. 
 Under such conditions 
 
 \l PQ = X 2 QQ = q, 
 
108 RADIOACTIVE TRANSFORMATIONS 
 
 where q is the number of atoms of emanation breaking up per 
 second. 
 
 The value of P, the number of atoms of radium B present 
 at any time t after removal from the emanation is given by 
 
 fo / P = 
 
 The value of Q is given by equation (1) as before. The solu- 
 tion of this equation is of the form 
 
 Q = a e-M + b 
 By substitution in equation (1) it is seen that 
 
 9 
 
 a = 
 
 Since initially when t = 0, Q = Q = , 
 
 AJJ 
 
 we have a + b = ; 
 
 A 2 
 
 thus h - ~" g A i 
 
 ' 
 
 and Q = * (e-M -**). (2) 
 
 A 2 AI A 2 
 
 The variation of the amount of radium B with time after 
 a long exposure is shown in Fig. 29, the number of atoms of 
 B initially present being 100. The number of atoms of radium 
 
 C present initially is jr-ZV 
 
 The curve, CC, expressing the number of atoms of C presen 
 at any time thus begins at a point whose ordinate is 77 insteac 
 of 100. 
 
 Since the ft or 7 ray activity of C is proportional at any time 
 to the value of $, the curve showing the variation of radium C 
 with time should be of the same form as the activity curve in 
 Fig. 27 for a long exposure, as measured by the {S and 7 rays 
 This is the case, for the theoretical and observed curves agree 
 within the, limit of experimental error. This is shown in the 
 following table: 
 
TRANSFORMATION OF RADIUM 
 
 189 
 
 DECAY or ACTIVITY MEASURED BY THE $ RAYS FOR A LONG EXPOSURE 
 TO THE EMANATION. 
 
 Time in minutes 
 after removal 
 from emanation. 
 
 Observed 
 activity. 
 
 Theoretical 
 activity. 
 
 
 
 100 
 
 100 
 
 10 
 
 97.1 
 
 
 96.8 
 
 20 
 
 88. 
 
 
 89.4 
 
 30 
 
 77., 
 
 
 78.6 
 
 40 
 
 67. 
 
 
 69.2 
 
 50 
 
 57.< 
 
 
 59.9 
 
 60 
 
 48.: 
 
 
 49.2 
 
 80 
 
 33. 
 
 
 34.2 
 
 100 
 
 22.i 
 
 
 22.7 
 
 120 
 
 14.5 
 
 14.9 
 
 /oo 
 
 60 
 
 40 
 
 2O 
 
 20 
 
 40 
 
 60 $O 
 
 IN MINUTES., 
 
 /OO 
 
 120, 
 
 FIG. 29. 
 
 Theoretical curves showing the number of atoms of radium B and radium C 
 existing at any time, when the matter initially consists of radium B and C in 
 radioactive equilibrium. 
 
110 
 
 RADIOACTIVE TRANSFORMATIONS 
 
 The fact that the long exposure curve shown in Fig. 27 re- 
 sults from two successive products, the first of which does not 
 emit rays at all, can readily be shown by graphical analysis. 
 
 Immediately after removal of the active body from the emana- 
 tion, the active deposit consists of B and C in equilibrium. 
 The /3 ray activity observed is due entirely to C, and, leaving 
 
 /oo 
 
 120 
 
 FIG. 30. 
 
 Analysis of the ft ray curve for a long exposure to the emanation, in order to 
 show that it results from the presence of two products, the first of which is 
 rayless. 
 
 out of account for the moment the fresh supply of C froi 
 the disintegration of B, the amount of C must, if left to itself, 
 diminish exponentially following the period of C, i. e., the 
 activity will fall to half value in 21 minutes. This decay 
 curve CC is shown in Fig. 30. Now the difference at any 
 time between the ordinates of the observed curve B + C, an< 
 
TRANSFORMATION OF RADIUM 111 
 
 the theoretical curve, CC, must be due to the activity of C, 
 supplied by the breaking up of B. This difference curve, BB 
 (see Fig. 30), should be identical in shape with the (S ray curve 
 of the active deposit for a short exposure. This must evidently 
 be the case, since this curve gives the activity arising from the 
 transformation of B alone, all the matter present initially being 
 of the kind B, which changes into C. 
 
 By comparison of the curve, BB, with the short exposure 
 curve shown in Fig. 27, this identity is seen to hold. The 
 activity rises from zero, and reaches a maximum after 35 
 minutes and then decays. 
 
 It is of interest to observe that the empirical equation of the 
 decay curve of the /3 ray activity for a long exposure to the 
 emanation was obtained before the theoretical explanation was 
 advanced. Curie and Danne 1 found that the activity I t at any 
 time could be expressed by an equation of the form 
 
 where \^ = 4.13 X 1Q- 4 (sec)- 1 and X 2 = 5.38 X 10-* (sec)- 1 , 
 
 and a = 4.20 is a numerical constant. The constant \ was de- 
 termined from the observed fact that the activity several hours 
 after removal from the emanation decayed exponentially with 
 a period of 28 minutes. The values of a and X 2 were deter- 
 mined so as to fit the curve. Now this equation is identical in 
 form with the theoretical equation for the activity when the first 
 change is rayless with a period of 28 minutes, and the second 
 change, which has a period of 21 minutes, gives out rays. 
 This is easily seen to be the case. From equation (2), the amount 
 Q of radium C, existing at any time , is given by 
 
 Ag AI Aj 
 
 Initially Q = Q = X z q. 
 
 1 Curie and Danne : Comptes rendus, cxxxvi, p. 364 (1903). 
 
112 RADIOACTIVE TRANSFORMATIONS 
 
 Since the activity at any time is proportional to the amount 
 of C present, i. e., to the value of , 
 
 It = Q_ 
 
 -*0 VO 
 
 On substituting the values of \ v A 2 , which correspond to 
 periods of 28 and 21 minutes respectively, 
 
 Xz = 4.3 and ^-_ = 3.3. 
 
 A 2 -A! A 2 - A, 
 
 Thus the theoretical equation not only agrees in form with 
 that deduced from observation, but the values of the constants 
 are very concordant. 
 
 Such a relation between theory and experiment would be 
 widely departed from if B as well as C gave out /3 rays. 
 
 ANALYSIS OF THE a RAY CURVES FOB A LONG EXPOSURE 
 
 We are now in a position to analyze the a ray activity curve 
 for a long exposure into its three components. In this case 
 we must take into account the first product, radium A, which 
 ^emits a rays. The observed a ray curve is shown in Fig. 31, 
 <iurve A + B + C. This curve was obtained by means of a 
 galvanometer. A piece of platinum foil was placed for several 
 days in a glass vessel containing a large supply of radium 
 emanation. The foil was then rapidly removed, placed on the 
 lower plate of a testing vessel, and a saturating voltage applied. 
 The variation of the a ray activity was measured by means of 
 a high resistance galvanometer. The initial value of the cur- 
 rent at the instant of removal was deduced by continuing the 
 curve backwards to meet the vertical axis. 
 
 The variation of activity with time is shown in the following 
 .table: 
 
TRANSFORMATION OF RADIUM 
 
 113 
 
 Time in 
 minutes. 
 
 Activity. 
 
 Time in 
 minutes. 
 
 Activity. 
 
 
 
 100 
 
 30 
 
 40.4 
 
 2 
 
 80 
 
 40 
 
 35.6 
 
 4 
 
 69.5 
 
 50 
 
 30.4 
 
 6 
 
 62.4 
 
 60 
 
 25.4 
 
 8 
 
 57.6 
 
 80 
 
 17.4 
 
 10 
 
 62.0 
 
 100 
 
 11.6 
 
 15 
 
 48.4 
 
 120 
 
 7.6 
 
 20 
 
 45.4 
 
 
 
 The activity due to A alone has almost vanished after 20 
 minutes. At the 20 minutes point the curve B -f C is produced 
 backwards, meeting the axis at L. It cuts it at about 50. The 
 difference between the ordinates of the curves A + B + C and 
 LL represents the activity due to radium A, and is shown by 
 
 IQO 
 
 60 
 
 o* 
 
 20 
 
 60 eo 
 
 IN MINUTES. 
 
 IQO 
 
 /no 
 
 FIG. 31. 
 
 Analysis of the a ray curve for a long exposure to the radium emanation, show- 
 ing that it results from the presence of three products, A, B, and C. Radium 
 A and C emit a rays, while radium B does not. 
 
 8 
 
114 RADIOACTIVE TRANSFORMATIONS 
 
 the curve AA, in the figure. The curve A is exponential with 
 a period of 3 minutes. The curve LL, B + C, is identical in 
 form over its whole range with the activity curve, for a long 
 exposure (see Fig. 27), measured by the ft rays. We may 
 conclude from this result that radium B emits no a rays. We 
 know already that it emits no /8 rays ; hence radium B must be 
 a rayless product. 
 
 The curve LL, B + C, can be analyzed into its two compo- 
 nents exactly in the same way as the corresponding /3 ray curve. 
 The curve CC represents the variation in the activity of the 
 radium C which existed at the moment of removal from the 
 emanation. The curve BB represents the activity due to C, sup- 
 plied by the change of B into C. This curve, BB, is identical 
 in form with the /3 ray curve for a short exposure (see Fig. 26). 
 
 We may conclude from this analysis that the active deposit 
 consists of three products, radium A, B, and C, which have 
 the following peculiarities : 
 
 Radium A emits only a rays, and is half transformed in 
 3 minutes. 
 
 Radium B is a rayless product, and is half transformed in 28 
 minutes. 
 
 Radium C emits a, /3, and 7 rays, and is half transformed in 
 21 minutes. 
 
 Several hours after removal from the emanation, the activity, 
 whether for a long or short exposure, and whether measured by 
 the a, /3, or 7 rays, falls off exponentially, with a period of 28 
 minutes. This is due to the fact that the longer period of the 
 rayless product, B, governs the final rate of decay, although 
 the activity is really supplied by the product C of 21 minute 
 period. 
 
 ARE RADIUM A AND B SUCCESSIVE PRODUCTS? 
 
 For simplicity, in the above comparison of theory with ex- 
 periment, the effect of radium A on the subsequent changes 
 has been neglected. If radium B is derived from radium A, 
 the amount of A present, when in radioactive equilibrium with 
 the emanation, is 3/28 or .11 of radium B. If A changes int 
 
 ' 
 
 ' 
 
 
TRANSFORMATION OF RADIUM 115 
 
 B according to a 3 minute period, the greater part of A will be 
 transformed in about 15 minutes, and it can be deduced that 
 the amount of B present after that interval should be about 
 8 per cent greater than if A did not change into B. The effect 
 of this on the subsequent decay curves should be easily measur- 
 able under suitable conditions. This point has been examined 
 by the writer, 1 but it was found that theory and experiment 
 agreed much more accurately if A and B were considered to 
 be independent products, separately produced during the trans- 
 formation of the emanation. The examination of the a ray 
 curve for a short exposure to the emanation does not give 
 definite evidence in either direction. 
 
 The conclusion that A and B are independent products, 
 however, involves such important theoretical consequences that 
 before accepting it, a close examination must be made to see if 
 the theoretical conditions are completely realized in practice. 
 
 The theory assumes that radium A should be deposited on 
 the electrode very shortly after its production, and that neither 
 A nor its subsequent products escape from the electrode ; or, in 
 other words, that these products show no appreciable volatility 
 at ordinary temperatures. 
 
 There is no doubt, however, that, under ordinary conditions, 
 appreciable quantities of both radium A and B, and sometimes C, 
 are present, mixed with the emanation, showing that all of these 
 products do not rapidly diffuse to the electrodes. In addition, 
 Miss Brooks 2 has shown that radium B is undoubtedly volatile 
 at ordinary temperatures. 
 
 Experiments are at present in progress in the laboratory of 
 the writer to decide whether such divergences between the theo- 
 retical and experimental conditions are sufficient to account for 
 the decay curves on the assumption that A and B are succes- 
 sive products. The question is still sub judice, but it is hoped 
 that a definite answer will soon be forthcoming. 3 
 
 1 Rutherford : Bakerian Lecture, Phil. Trans. A, p. 169, 1904. 
 
 2 Miss Brooks : Nature, July 21, 1904. 
 
 8 The cause of this discrepancy between theory and experiment has been indicated 
 by some recent experiments of H. W. Schmidt (Physik. Zeit., 6, No. 25, p. 897, 1905). 
 
 I 
 
116 RADIOACTIVE TRANSFORMATIONS 
 
 If radium A and B are proved to be independent, it will be 
 necessary to suppose that the emanation breaks up into two dis- 
 tinct products, and also expels one or more a particles. The 
 observed fact that the activity due to radium A for a long ex- 
 posure is nearly equal to that of radium C when the plates are 
 close together, is in agreement with both hypotheses, provided 
 that it is assumed in the non-successive hypothesis that each 
 atom of the emanation breaks into two products, besides expel- 
 ling an a particle. 
 
 On such a view, the emanation gives rise to two distinct 
 families of products. While a possible change of this character 
 is of interest and importance, evidence of an undeniable char- 
 acter must be forthcoming before it can be accepted. If a 
 product can be separated from radium or its active deposit, 
 which decays exponentially, with a 3 minute period, and does 
 not give rise to radium B and C, the independence of A and B 
 will be completely established. 
 
 EFFECT OF TEMPERATURE ON THE ACTIVE DEPOSIT 
 
 It has been assumed without proof in the above discussion 
 that radium B rather than radium C has the period of 28 minutes. 
 The comparison of theory with experiment does not throw any 
 light on the question, since the activity curves are the same if 
 the periods of the products B and C are interchanged. 
 
 As in the case of thorium, it is necessary to have recourse 
 to other evidence to settle whether the period of 28 minutes 
 belongs to B or C. It is necessary by some physical or chemi- 
 cal means to isolate B from C, and to examine separately their 
 rates of change. 
 
 This has been effected by taking advantage of the greater 
 
 He finds that radium B is not a rayless product, but emits /3 rays which are of much 
 smaller penetrating power than those from radium C. We have seen (Fig. 26) that 
 the ray curve for a short exposure to the emanation passes through a maximum 
 35 minutes after removal. This only holds when the rays have been passed through 
 a screen of sufficient thickness to absorb the rays from radium B. With thinner 
 screens, the maximum is reached earlier. When this new factor is fully taken into 
 account, it appears probable that the experimental curves will completely agree with 
 the theory that radium A, B, and C are successive products. 
 
TRANSFORMATION OF RADIUM 11T 
 
 volatility of radium B when an active wire is exposed to a 
 high temperature. Miss Gates l observed that the active de- 
 posit of radium was volatilized at a white heat, and redeposited 
 on the cold bodies in the neighborhood. Curie and Danne 2 
 examined this effect in more detail, and obtained some very 
 interesting results. An active wire surrounded by a cold metal 
 cylinder was heated for a short time by an electric current, and 
 the activity both of the wire itself and of the interior of the 
 cylinder separately examined. At about 400 C., some of the 
 radium B was volatilized. This was established by noting 
 the variation of the activity of the distilled part after the heat- 
 ing. This activity was small at first, passed through a maxi- 
 mum, and then decayed, in exactly the same way as the ft ray 
 activity of a body for a short exposure (see Fig. 26). This 
 showed that the matter initially deposited on the cylinder con- 
 sisted only of the rayless product B, which changed into the ray 
 product C. At a temperature of about 600 C. most of the B was 
 driven off, and also some C. 
 
 A number of experiments were then made on the decay of 
 activity of the wire for temperatures varying between 15C. 
 and 1350C. At a temperature of 630C., they state that the 
 activity of the wire decreased exponentially with a period of 28 
 minutes. The period steadily decreased to 20 minutes, while 
 the temperature rose to 1100 C. At this temperature it passed 
 through a minimum value, and then increased to 25 minutes 
 at 1300C. 
 
 Since their decay curves were exponential, Curie and Danne 
 supposed that all of B was volatilized at 630 C. If this were 
 the case, the results indicated that the 28 minute period must 
 be ascribed to radium C and the 21 minute period to^radium B. 
 The experiments also indicated that rise of temperature above 
 630 C. to 1100C. produced an apparent alteration in the rate 
 of change of radium C. This, if correct, was a most important 
 result, for there was no previous evidence that temperature had 
 any effect in altering the rate of transformation of a radioactive 
 
 1 Miss Gates : Phys. Rev., p. 300, 1903. 
 
 2 Curie and Danne : Comptes rendus, cxxxviii, p. 748, 1904. 
 
118 RADIOACTIVE TRANSFORMATIONS 
 
 product. According to their experiments, the rate of transfor- 
 mation of radium C was altered in an unexpected manner by rise 
 of temperature, for it increased up to 1100C. and at a still 
 higher temperature fell again nearly to its normal value. 
 
 A close examination of the effect of temperature on the 
 active deposit was recently made by Dr. Bronson 1 in the 
 laboratory of the writer. The results obtained by him showed 
 conclusively that a rise of temperature to 1100 C. has no effect 
 in altering the rate of transformation of the active deposit, and 
 that the experimental results obtained by Curie and Danne can 
 be explained by supposing that in most of their experiments the 
 deposit on the wire after heating consisted not entirely of C but 
 of a mixture of B and C. 
 
 In order to test definitely whether temperature has any effect 
 on the decay of the active deposit, an active copper wire was 
 placed in a small length of combustion tubing, and the glass 
 sealed under diminished pressure. This was then heated in an 
 electric furnace to different temperatures. The glass was found 
 to withstand a temperature of about 1100C. The /3 ray activ- 
 ity was then carefully examined over a long interval. Between 
 2.5 and 4 hours, the curves obtained were approximately ex- 
 ponential with a period of 28 minutes. After 6 hours, the curve 
 was accurately exponential with a period of about 26 minutes. 
 Within the limit of experimental error, the curves of decay up 
 to 1100 C were found to be identical with the normal decay 
 curves at atmospheric temperature. 
 
 In this experiment, none of the distilled products were able 
 to escape, so that we may conclude with certainty that a rise of 
 temperature up to 1100C. has no appreciable effect on the rate 
 of transformation of the active products. 
 
 On repeating the experiments of Curie and Danne, Bronson 
 found that the decay of activity after heating the wire to a con- 
 stant temperature was very variable, and that approximately 
 exponential curves were obtained with periods lying between 
 25 and 19 minutes after heating the wire to the same tempera- 
 ture. If, for example, a current of air was blown through the 
 
 1 Bronson : Amer. Journ. Sci., July, 1905. 
 
TRANSFORMATION OF RADIUM 119 
 
 electric furnace before removing the wire, the activity fell ex- 
 ponentially with a period of about 19 minutes. In a similar way 
 if a cold copper wire was introduced above the active wire the 
 period had about the same value. In such cases, there is a better 
 opportunity for the distilled part, radium B, to escape from the 
 wire. Several curves were obtained which were accurately ex- 
 ponential with a period of 19 minutes. This result showed 
 that the active product, radium C, has a period of 19 minutes, 
 and that the 26 minute period must be ascribed to radium B. 
 
 It was observed that in all cases in which the activity de- 
 cayed initially with a period lying between 19 and 26 minutes, 
 the curves were not at first accurately exponential. The period 
 always tended towards a value of 26 minutes as the activity de- 
 cayed, and the law was then exponential. This is exactly what 
 is to be expected if the activity after heating the wire results 
 from a mixture of B and C, the amount of C initially predomi- 
 nating. The activity will first decay to half value with a period 
 intermediate between those of B and C. The amount of radium 
 B, which changes more slowly than C, after a time begins to pre- 
 dominate, and ultimately governs the final rate of decay, i. e., 
 the activity falls finally according to an exponential law with a 
 period of 26 minutes. 
 
 The experiments have thus shown that while B is more vola- 
 tile than C, in many cases all of the B is not removed even if 
 the wire is heated to a temperature far above its point of 
 volatilization. 
 
 The periods of the two products, radium B and C, are 26 and 
 19 minutes respectively, values somewhat lower than the periods 
 28 and 21 minutes assumed in the previous calculations. Be- 
 tween 2 and 4 hours after removal from the emanation, the 
 activity under normal conditions falls approximately exponen- 
 tially with a period of 28 minutes, and this originally led to the 
 choice of 28 minutes as the value of one of the periods. The 
 decay curve, however, is not accurately exponential until about 
 6 hours after removal from the emanation, and the period is then 
 26 minutes. 
 
 In the analysis of the changes, radium C has been given the 
 
120 RADIOACTIVE TRANSFORMATIONS 
 
 shorter period, but the original determinations of the periods 
 of B and C viz., 28 and 21 minutes, have been retained. Over 
 the range considered, the theoretical curves for periods of B 
 and C of 26 and 19 minutes respectively do not differ much 
 from those with the periods of 28 and 21 minutes. 
 
 The retention of the old values brings out more clearly the 
 methods originally employed of proving that B was a rayless 
 product and that C gives out a, /3, and 7 rays. A more accu- 
 rate determination of the various curves of decay during the 
 first two hours is at present in progress. 
 
 The series of transformation products of radium which have 
 so far been discussed are diagrammatically shown in Fig. 32. 
 
 ACTIVE oefos/i 
 FIG. 32. 
 Radium and its family of rapidly changing products. 
 
 The periods of the products are given for convenience, as well 
 as the character of the emitted rays. 
 
 It is a matter of remark that of these five radioactive sub- 
 stances, only radium C gives out ft and 7 rays. The others 
 emit only a rays. The emission of the a rays is, however, 
 accompanied by a secondary radiation, which is produced prob- 
 ably by the impact of the a rays on matter and consists of 
 electrons which are projected at a speed small compared witl\ 
 that of the ft rays proper, and which are consequently very 
 easily deflected by a magnetic field. The presence of such slow 
 moving electrons was first observed by J. J. Thomson l for 
 radiotellurium and by Rutherford 2 for radium. 
 
 1 J. J. Thomson : Proc. Camb. Phil. Soc., Nov. 14, 1904. 
 
 2 Rutherford: Phil. Mag., Aug., 1905. 
 
TRANSFORMATION OF RADIUM 121 
 
 Miss Slater l has recently shown that the emission of a par- 
 ticles from the thorium and radium emanations is also accom- 
 panied by slow moving electrons carrying a negative charge. 
 
 The expulsion of such electrons is probably not a true radia- 
 tion from the active matter itself, but is largely a secondary 
 effect produced when the a particles impinge ,upon or escape 
 from matter. It is for this reason not advisable to call them 
 ft rays, for this name should be retained for the primary ft par- 
 ticles emitted from radioactive substances with velocities ap- 
 proaching that of light. J. J. Thomson has suggested that the 
 name 8 rays be applied to such slowly expelled electrons. 
 
 In the next chapter, we shall show that the changes in 
 radium do not end with radium C, but continue through three 
 more distinct stages. The calculations adopted to analyze the 
 active deposit of rapid transformation are not, however, appre- 
 ciably affected by the presence of these further products, for 
 the activity due to them is in most cases less than one millionth 
 of that observed on the active body immediately after removal 
 from the emanation. 
 
 1 Miss Slater: Phil. Mag., Oct., 1905. 
 
CHAPTER V 
 
 ACTIVE DEPOSIT OF RADIUM OF SLOW 
 TRANSFORMATION 
 
 A BODY which has been exposed in the presence of the 
 radium emanation and then removed does not completely lose 
 its activity. A small residual activity is always observed, the 
 amount depending not only on the quantity of the emanation 
 to which it has been exposed, but also upon the time of expo- 
 sure. This small residual activity was first observed by Mme. 
 Curie and has been closely examined by the writer. 
 
 After removal from the emanation, the activity of a body at 
 first decays according to the laws discussed in the last chapter. 
 There is finally an exponential rate of decay with a period of 
 26 minutes. Twenty-four hours after removal, the active de- 
 posit of rapid transformation has disappeared almost completely, 
 and the activity left behind is generally less than one millionth 
 of the activity observed immediately after removal from the 
 emanation. 
 
 In the present chapter, the variations of this activity with 
 time will be considered and the changes occurring in the matter 
 deduced. The active deposit of slow transformation will be 
 shown to consist of three successive products called radium 
 D, E, and F. The analysis of this apparently insignificant 
 residual activity observed on bodies has yielded results of con- 
 siderable importance. It has disclosed the origin of the radio- 
 lead of Hofmann, of the radiotellurium of Marckwald, and 
 also of the polonium of Mme. Curie; for these substances will 
 be shown to be derived from the transformation of the radium 
 atom. 
 
 It might at first sight be thought that this slight residual 
 activity observed in bodies was due, not to the deposit of an 
 
SLOW TRANSFORMATIONS OF RADIUM 123 
 
 active substance upon them, but to a possible effect produced 
 by the powerful radiations of the emanation to which the bodies 
 had been exposed. 
 
 This point was examined by the writer 1 in the following 
 way: 
 
 The interior surface of a glass tube was covered with equal 
 areas of thin metal, including platinum, aluminium, iron, copper, 
 silver, and lead. A large quantity of emanation was intro- 
 duced into the tube and left there for seven days. The activi- 
 ties of the plates, two days after removal from the emanation, 
 were separately tested, and found to be unequal, being greatest 
 for copper and silver and least for aluminium. 
 
 After standing for another week, the initial variations of 
 activity had largely disappeared. These initial differences of 
 activity were due to slight differences in the rates of absorption 
 of the radium emanation by the metals. As this emanation was 
 released, the activities of the plates reached equal values. The 
 radiations from each plate consisted of a and @ rays and were 
 identical in penetrating power. This result shows that the 
 residual activity observed cannot be due to any direct actions 
 of the radiations on the body, for if this were the case, we 
 should expect the activity of the different metals to vary not 
 only in quantity but in quality. We may conclude then, that 
 the activity is due to an active substance deposited on the 
 surface of the metals. This view is completely borne out by 
 later experiments, for it will be shown that the active deposit 
 can be removed from a platinum plate by solution in acids and 
 can also be volatilized at a high temperature. 
 
 VAKIATION OF THE a RAY ACTIVITY WITH TIME 
 
 The a ray activity of the body after reaching a minimum 
 value during the first few days, steadily increases in amount 
 for several years. The activity during the first few months 
 increases nearly proportionally with the time. The curve 
 (Fig. 33) then begins to bend over, and after 240 days the 
 period over which it has so far been examined becomes much 
 
 i Rutherford : Phil. Mag., Nov., 1904. 
 
124 
 
 RADIOACTIVE TRANSFORMATIONS 
 
 more flattened and obviously approaches a maximum value. The 
 explanation of this rise of activity will be considered at a later 
 stage. 
 
 VARIATION OF THE ft RAY ACTIVITY WITH TIME 
 
 The residual activity initially comprises both a and ft rays, 
 the latter being present in a much greater relative proportion 
 
 FIG. 33. 
 
 Rise of a ray activity of a body after exposure to the radium emanation, 
 activity is a measure of the amount of radium F present. 
 
 The 
 
 than is observed in radium or uranium. The ft ray activity is 
 small at first, but increases with time, reaching a maximum 
 after about 50 days. The variation of activity with time is 
 shown in Fig. 34. A plate was exposed for 3.75 days to the 
 radium emanation, and the observations of the ft ray activity, 
 by means of an electroscope, were begun 24 hours after remo- 
 val. The time was measured from the middle of the period of 
 exposure to the emanation. The curve is seen to be similar 
 
SLOW TRANSFORMATIONS OF RADIUM 125 
 
 in shape to the recovery curve of the emanation or of ThX. 
 The /3 ray activity I t at any time t after removal is given by 
 
 The activity reaches half value in about 6 days, and after 50 
 days has nearly reached a maximum. 
 
 J 
 
 ,5 /o is s.o -90 oo 
 TIME: IN DAYS 
 
 FIG. 34. 
 
 Rise of $ ray activity of a body after exposure to the radium emanation. The 
 ft ray activity is a measure of the amount of radium E present. 
 
 Observations of the fi ray activity were continued for 18 
 months, but showed that the activity remained practically con- 
 stant after 50 days. 
 
 A curve of this character indicates that the /3 ray product 
 is produced at a constant rate from a primary source, whose 
 rate of transformation is so slow as to appear nearly constant 
 over the period of observation. It follows from the curve that 
 the {3 ray product has a period of 6 days. 
 
 The fact that the a and fj ray activities increase almost from 
 
126 RADIOACTIVE TRANSFORMATIONS 
 
 zero shows that their primary source, called radium D, is a ray- 
 less product, which, as we shall see later, is probably half trans- 
 formed in about 40 years. Radium D is transformed into the 
 ft ray product called radium E, which is half transformed in 
 about 6 days. 
 
 EFFECT OF TEMPERATURE ON THE ACTIVITY 
 
 A platinum plate several months after its removal from the 
 emanation was placed in an electric furnace and heated for a few 
 minutes to varying temperatures. Exposure for four minutes, 
 first at 430C. and later at 800C., had little if any effect in 
 altering either the a or ft ray activity. The a ray activity was, 
 however, almost completely removed by heating the plate to 
 about 1050 C., while the ft ray activity did not at the time show 
 any change. This result shows clearly that the product which 
 emits a rays is more volatile than the product which emits 
 ft rays. 
 
 This experiment is another example of the way in which 
 differences in volatility of two products may be utilized to 
 effect a partial separation of one from the other. 
 
 We now come to another striking observation. The ft ray 
 activity of the platinum plate, though apparently unchanged 
 immediately after the heating, began slowly to decrease, and 
 finally fell to one quarter of the initial value. Subtracting this 
 residual activity, it was found that the ft ray product lost its 
 activity exponentially, falling to half value in about 4.5 days. 
 
 We may thus conclude that the heating of the active deposit 
 had a double action ; for not only was the a ray product (which 
 will be shown to arise from the ft ray product, radium E) driven 
 off, but about three quarters of the primary source, radium D, 
 was also volatilized. 
 
 We thus have the striking result that in a mixture of three 
 successive products, the first and third are mostly volatilized 
 at a temperature of about 1000 C., while the middle product is 
 unaffected. It will be observed that the period of decay of the 
 ft ray product (4.5 days), observed after heating, does not agree 
 with the period of the same product (6 days) deduced from the 
 
SLOW TRANSFORMATIONS OF RADIUM 12T 
 
 rise curve of Fig. 33. This difference requires further inves- 
 tigation. The 6 day period is probably the more correct value 
 under normal conditions. 
 
 SEPARATION OF THE a RAY PRODUCT BY BISMUTH 
 
 The emanation from 30 milligrams of radium bromide was 
 condensed in a glass tube and left there for one month. The 
 active deposit remaining on the surface of the glass was then 
 dissolved in dilute sulphuric acid and the solution laid by for 
 about a year. During this interval the a ray activity steadily 
 increased. By introducing a polished bismuth disc into the 
 solution, the a ray product may be deposited electrochemically 
 on the bismuth. By introducing several bismuth discs suc- 
 cessively into the solution and allowing them to remain for 
 several hours, the a ray product was mostly removed. On 
 evaporating the solution to dry ness, it was found that only 10 
 per cent of the original a ray activity remained behind. 
 
 The @ ray activity of the solution was not altered by this 
 process. The bismuth discs gave out only a rays, but no trace 
 of /3 rays. This result shows that only the a ray product was 
 removed. Radium D, as well as E, was left behind, for if some 
 radium D had been deposited on the bismuth, it would have 
 changed into radium E, and consequently some ft rays would 
 have been emitted from the bismuth disc, after standing for 
 several weeks in order to allow time for D to change into E. 
 
 No such effect, however, was observed. The activities of 
 these bismuth discs were tested in an a ray electroscope for 
 over 200 days. The activity of each was found to fall off 
 nearly exponentially, reaching half of the initial value after 
 about 143 days. We may thus conclude that the substance 
 which emits a rays is a simple product, which is half trans- 
 formed in 143 days. This a ray product will be called radium 
 F, for it will be shown to be a successive product of radium E. 
 
 The fact that radium E is the parent of F is clearly brought 
 out by the following experiment. 
 
 A platinum wire coated with the active deposit of slow 
 transformation was exposed for some minutes to a temperature 
 
128 
 
 RADIOACTIVE TRANSFORMATIONS 
 
 of over 1000 C. Most of the radium F was volatilized. The a 
 ray activity of this platinum plate was then carefully examined 
 for several weeks. The small a ray activity, observed imme- 
 diately after removal from the furnace, increased rapidly during 
 the first two weeks and then more slowly. The gain of a ray 
 activity with time is shown in Fig. 35. 
 
 /COj 
 
 60 
 
 80 
 
 - TIMS IN OAYQ - 
 
 FIG. 35. 
 
 Hise of a ray activity on a platinum plate which has been heated to a tempera- 
 ture sufficient to remove most of the radium D and F present. Radium E is 
 left behind and changes into F. 
 
 A curve of this character is to be expected if radium E is the 
 parent of F. The action of the high temperature volatilized 
 the greater part of D and F, but left E behind. The radium 
 E was then transformed with a period of 4.5 days and changed 
 into F. The a ray activity thus initially rose rapidly, due to 
 the new radium F supplied by the transformation of E. The 
 slow increase observed after some weeks, when most of the 
 radium E present had been transformed, was due to the pro- 
 duction of radium F by the small amount of D and E which 
 was not volatilized from the platinum plate. 
 
 We may thus conclude that radium E is the parent of 
 radium F. 
 
SLOW TRANSFORMATIONS OF RADIUM 129 
 
 It has been previously shown that radium E is produced from 
 radium D, which itself does not emit /3 rays. The small a ray 
 activity observed initially when radium D is present in maxi- 
 mum amount shows that radium D does not emit a rays. 
 Radium D is consequently a rayless product. 
 
 SUMMARY OF THE RADIUM PRODUCTS 
 
 The analysis of the active deposit of slow change has thus 
 disclosed the existence of three successive products of radium. 
 The period of change of these products and some of their dis- 
 tinctive physical and chemical properties are tabulated below. 
 
 Product. 
 
 Time to be half 
 transformed. 
 
 Rays. 
 
 Chemical and physical properties. 
 
 Radium D 
 
 About 40 years 
 
 None 
 
 Soluble in strong acids, volatilized 
 
 
 
 
 at or below 1000 C. 
 
 " E 
 
 6 days 
 
 $ and (7?) 
 
 Non-volatile at 1000 C. 
 
 F 
 
 143 
 
 a 
 
 Volatile at about 1000 C., depos- 
 
 
 
 
 ited from a solution on a bis- 
 
 
 
 
 muth plate. 
 
 The method for deducing the period of radium D will be con- 
 sidered a little later. A sufficient amount of radium E has 
 not yet been collected to test whether it gives out 7 as well as 
 /3 rays. But since in every other substance examined, these 
 two types of rays always go together, it is almost certain that 
 radium E gives out 7 rays. 
 
 In the last chapter it was shown that the active deposit of 
 rapid change consists of the three successive products, radium 
 A, B, and C. It is thus natural to conclude that radium D is 
 derived directly from the transformation of radium C. It is 
 difficult to show definitely that radium C is the parent of D. 
 We know, however, that D must be derived from either the 
 emanation or one of its products, and since the products A, B, 
 and C are lineal descendants of the emanation, the most plausi- 
 ble assumption is that the family of products D, E, and F are 
 also lineal descendants of radium C. 
 
 On this assumption, the various radium products, together 
 
 9 
 
130 RADIOACTIVE TRANSFORMATIONS 
 
 with their periods of transformation and the types of rays 
 emitted, are shown diagrammatically below (Fig. 36). 
 
 It is instructive for a moment to review briefly the series of 
 changes exhibited by radium. The radium atom is a compara- 
 tively stable one, and on an average only half the radium atoms 
 break up in 1300 years. The a particle projected during the 
 disintegration of the atom has a velocity slower than those from 
 the radium products, and can only pass through 3. 5 cms. of air 
 before complete absorption. The radium suffers a radical change 
 on account of the loss of the a particle, and is transformed into 
 a gas the radium emanation which is far more unstable 
 than radium itself, for half of it breaks up in 3.8 days. After 
 the expulsion of an a particle, the product radium A makes 
 
 > "* <i ^ v 
 
 <5-d-6-o- o- o- c 6- 
 
 EMAN. /?AA f^AaB RAD.C RADD 
 
 iJhJirjS. 
 
 A */o CHANGE ACTIV 
 
 FIG. 36. 
 Radium and its family of products. 
 
 its appearance. This is the most unstable of all the radium 
 products, for half of it breaks up in 3 minutes. 
 
 The next product is radium B, with a period of 26 minutes. 
 It has the peculiarity of being transformed without the emission 
 of rays at all. This points either to a transformation by the 
 rearrangement of the components of the atom without any loss 
 of mass, or, as is more likely, to the emission of an a particle at 
 a velocity too low to ionize a gas. It will be seen later that the 
 a particle loses the property of ionization when its velocity falls 
 below about one fortieth of the velocity of light, so that the a 
 particle may be expelled at a considerable speed and yet show 
 no ionization effects. The next substance is radium C, which 
 is the most remarkable of all the radium products, for in 
 breaking up it emits all three types of rays. It would appear 
 
SLOW TRANSFORMATIONS OF RADIUM 131 
 
 as if the transformation of C were accompanied by a most 
 violent explosion in the atom, for not only is the a particle 
 ejected with a greater speed than from any of the other prod- 
 ucts, but at the same time @ particles are expelled with a 
 velocity nearly equal to that of light. There is also an emission 
 of very penetrating 7 rays. 
 
 The a particle projected from radium C can traverse 7 cms. 
 of air before complete absorption, while the a rays from the 
 other products have a range not greater than 4.8 cms. After 
 this violent atomic outburst, the residual atom, radium D, is far 
 more stable and breaks up without the appearance of rays. 
 
 The next product is radium E, which emits only @ and y 
 rays. It has a comparatively short life, but gives rise in turn to 
 radium F, which has a slow period of change. No further prod- 
 ucts of transformation have been detected, and the interesting 
 question of the final or end product of radium is reserved for 
 discussion in Chapter VIII. 
 
 PERIOD OF CHANGE OF RADIUM D 
 
 Radium D does not emit rays, and consequently neither its 
 properties nor its rate of transformation can be determined by 
 direct means. The following product, radium E, however, 
 emits y3 rays, and by noting the variations of its activity when 
 in equilibrium we should be able to detect any variation in the 
 rate of change of the parent product D. 
 
 This is readily seen to be the case, for when equilibrium is 
 reached, the number of atoms of E which break up per second 
 will always be equal to the number of atoms of D breaking up 
 per second. Unfortunately the rate of transformation of D is 
 so slow that no certain change in the equilibrium activity of E 
 has been detected in the course of one year, and a long interval 
 of time will probably be necessary to fix the period of D by 
 direct measurement. 
 
 It is of importance to form a rough estimate of its probable 
 period. This can be deduced on certain assumptions which 
 are probably approximately realized in practice. 
 
 Suppose that a quantity of emanation is introduced into a 
 
132 RADIOACTIVE TRANSFORMATIONS 
 
 closed vessel. and left there to decay. Several hours after the 
 introduction of the emanation, the amount of radium C which 
 emits ft rays reaches a maximum value, and then decays at the 
 same rate as the emanation. If q l is the maximum number of 
 ft particles emitted per second from radium C at its maximum 
 activity, the total number -ZVj emitted during the life of the ema- 
 
 nation is very approximately given by -ZVi = , where X x is the 
 
 A! 
 
 constant of change of the emanation. Suppose that the active 
 deposit of slow transformation is allowed to remain undisturbed 
 for about 50 days after the emanation has practically disappeared. 
 Radium D and E will then be in equilibrium. Let q 2 be the 
 number of ft particles emitted from D and E. Then if the 
 transformation of radium D follows the ordinary exponential 
 law with a constant X 2 , the total number N 2 of ft particles 
 emitted during the life of radium D is given as before by 
 
 N 2 = . But if each atom of radium C in breaking up 
 A 2 
 
 emits one ft particle, the total number of ft particles emitted 
 during the life of the emanation must be equal to the number 
 of atoms of emanation originally present. The number of atoms 
 of D formed by the emanation will also be equal to this quan- 
 tity, and if each atom of D gives rise to one of E, which breaks 
 up with the expulsion of one ft particle, we see that the total 
 number of ft particles expelled from C during the life of the 
 emanation must be equal to the total number of particles 
 expelled from E during the life of D. Consequently JV^ = -ZV 2 > 
 and therefore 
 
 It is not easy to measure directly the number of ft particles 
 expelled either from radium C or E, but on the assumption that 
 the average ft particle emitted from C or E produces about the 
 same ionization in a gas, 
 
SLOW TRANSFORMATIONS OF RADIUM 133 
 
 where ?\, i z are the saturation ionization currents due to C and E 
 respectively, measured under the same conditions in the same 
 
 testing vessel. This ratio, -^, can be readily determined, so that 
 
 A ^ 1 
 
 the ratio -2 is known. Substituting the value of X x for the 
 
 emanation, X 2 can be determined. 
 
 Proceeding by this method, the writer 1 deduced that radium 
 D should be half transformed in about 40 years. This period 
 is almost certainly of the right order, but from the nature of the 
 assumptions, the value cannot pretend to be more than a first 
 approximation to the truth. The main source of error probably 
 lies in the assumption that the /3 particles of radium C and E 
 produce the same average ionization in the gas. 
 
 As a criterion of the order of accuracy obtained in predicting 
 periods of change by these means, it may be mentioned that, by 
 a similar method, I deduced that the period of radium F was 
 about one year. Actual observation has since shown that this 
 period is 143 days. I think that the period of D will certainly 
 be found to lie between 20 and 80 years. 
 
 VARIATION OF THE a AND RAY ACTIVITY OVER LONG 
 PERIODS OF TIME 
 
 We are now in a position to deduce the variation of the a 
 and ft ray activity for the active deposit over long intervals of 
 time. Since radium E is transformed at a rapid rate compared 
 with F, we may assume as a first approximation that D is trans- 
 formed directly into F. The problem thus reduces to the fol- 
 lowing: Given that the periods of two successive products are 
 40 years and 143 days respectively, find the number of atoms 
 of each product present at any time. This is exactly equivalent 
 to the practical case already considered on page 50 for the active 
 deposit of thorium where the two changes had periods of 
 11 hours and 55 minutes respectively. 
 
 The ft ray activity of D and E after reaching its maximum 
 will decrease exponentially, falling to half value in 40 years. 
 
 1 Kutherford, Phil. Mag., Nov., 1904. 
 
134 
 
 RADIOACTIVE TRANSFORMATIONS 
 
 Using the equation discussed on page 51, it can at once be 
 deduced that the number of atoms of radium F reaches its maxi- 
 mum in about 2.6 years and that this substance will ultimately 
 decay pari passu with the parent product D, i. e., it will be 
 half transformed about 40 years later. The curves shown in 
 Fig. 37 give the relative number of atoms of E and F which 
 break up per second at any time after the formation of the 
 
 100 ^_ 
 
 fiftj 
 
 E 
 
 
 
 
 
 
 
 - . 
 
 ^s-* 
 
 
 
 = r= 
 
 === 
 
 === 
 
 === 
 
 60J / 
 
 
 
 
 
 
 
 
 J/ 
 
 
 
 
 
 
 
 
 Jf 
 
 
 
 
 
 
 
 
 L 
 
 
 
 - lYtfitf 
 
 }r<rara 
 
 
 
 
 1 2345678 
 
 FIG. 37. 
 
 Curve E E represents the variation in the number of atoms of radium D break- 
 ing up per second. Curve F F represents the number of atoms of radium F 
 breaking up per second. 
 
 deposit. Since the activity of F is proportional to the number 
 of atoms of F which break up per second, we see that the activ- 
 ity of F will rise from zero to a maximum after 2.6 years, and 
 will then decay with a 40 year period. 
 
 The variation of the a ray activity with time, is in good 
 agreement with the theoretical curve over the range so far 
 examined (See Fig. 33). 
 
 It is of interest to note that the same a ray activity observed 
 
SLOW TRANSFORMATIONS OF RADIUM 135 
 
 9 days after the formation of the active deposit is again reached 
 after an interval of about 180 years. 
 
 The production by radium of the active deposit of slow trans- 
 formation at once explains the strong radioactivity observed in 
 rooms in which large quantities of radium have been used, even 
 when the radium has been completely removed for some time. 
 This effect has been observed by several experimenters, and 
 especially by those who have been occupied in the separation 
 and concentration of large quantities of radium. 
 
 The radium emanation released from the radium is trans- 
 ferred by diffusion and convection currents throughout the 
 whole laboratory, and far distant rooms into which radium prep- 
 arations have not been introduced become permanently radio- 
 active. The emanation is transformed in situ through the 
 succession of products radium A, B, and C, and finally passes 
 into the active deposits of slow change. This matter is de- 
 posited on the interior surface of rooms and on every object in 
 the building. For a given supply of emanation, the a ray 
 activity will be small at first, but will steadily increase for about 
 three years. 
 
 This residual activity on bodies is a source of considerable 
 disturbance in radioactive work. Eve, 1 for example, found 
 that every substance examined in the Macdonald Physical 
 Laboratory of McGill University showed an abnormally large 
 natural activity. At the time of testing, this activity was 
 about 60 times greater than that observed in the same labora- 
 tory before the introduction of large quantities of radium into 
 the building. All electroscopes made of materials exposed in 
 the building had a large natural leak due to the active deposit. 
 This can in part be removed by cleaning with sandpaper or by 
 solution in acids. Unless all electroscopes or testing vessels 
 are made outside the laboratory to insure a small natural leak, 
 measurement of very weak radioactivities is rendered almost 
 impossible. When once a building has been infected, it does 
 not serve any immediate purpose to remove the radium, for the 
 
 i Eve: Nature, March 16, 1905. 
 
136 RADIOACTIVE TRANSFORMATIONS 
 
 a ray activity will continue to increase for about three years 
 and will last for hundreds of years. 
 
 For these reasons it is very advisable to reduce the escape of 
 emanation into the air of a laboratory as far as possible and to 
 keep all radium salts in sealed vessels. 
 
 PRESENCE OF THE ACTIVE DEPOSIT IN RADIUM 
 
 Since radium D is produced from radium at a nearly constant 
 rate, it should gradually increase in quantity with the age of 
 the radium. The presence of radium D in old radium can be 
 detected in a very simple way. With a freshly prepared sample 
 of radium, continued boiling for five or six hours removes the 
 emanation as fast as it is formed, and reduces the /3 ray activity 
 arising from radium C to a fraction of one per cent of its 
 maximum value when in radioactive equilibrium. 
 
 A very different result is observed if an old preparation of 
 radium is treated in a similar way. The writer had in his 
 possession a small quantity of impure radium kindly presented 
 by Professors Elster and Geitel four years before. 
 
 After continued boiling, the activity could not be reduced 
 below 8 per cent of the original amount, or about 9 per cent of 
 the activity due to C alone. This residual /3 ray activity was 
 due to the radium E stored up in the compound. 
 
 The amount of radium E will steadily increase with time, and 
 will reach a maximum value when the same number of atoms of 
 radium C and radium E break up per second. The number of 
 /B particles expelled per second from radium E will, under such 
 conditions, be equal to the number expelled from radium C. 
 Since the radium itself is transformed very slowly compared with 
 radium D, the amount of radium D produced per year (measured 
 by the y5 ray activity of radium E) should be about 1.7 per cent 
 of the equilibrium amount. 
 
 The ray activity due to radium E should thus be about 
 7 per cent of that due to radium C after the lapse of four years. 
 The observed and calculated values (9 and 7 per cent respec- 
 tively) are thus in fair agreement. 
 
 By adding a trace of sulphuric acid to the radium solution, 
 
SLOW TRANSFORMATIONS OF RADIUM 137 
 
 the radium was precipitated and the products D, E, F, which 
 are soluble in sulphuric acid, remained in the solution. The 
 filtrates thus contained a large part of the above three products. 
 The radium F was removed from the solution by means of 
 bismuth discs, and showed an activity to be expected from the 
 age of the radium. 
 
 VARIATION OF THE ACTIVITY OF RADIUM WITH TIME 
 
 We shall see later that radium itself, apart from its products, 
 is probably half transformed in about 1300 years, and conse- 
 quently the number of atoms breaking up per second decreases 
 exponentially according to this period. In consequence of the 
 formation of the active deposit of slow change, the activity 
 supplied by it will at first more than compensate for the de- 
 crease in the activity of the radium itself. The activity will 
 rise for several hundred years, but will ultimately decay expo- 
 nentially with the period of the radium. 
 
 When sufficient time has elapsed for approximate equilibrium 
 between the mixture of products, the number of /3 particles 
 expelled from the old radium will be twice that due to radium 
 C alone; for radium E will emit per second, under such con- 
 ditions, the same number of /3 particles as radium C. 
 
 It can readily be calculated from the theory of two changes, 
 that the number of $ particles expelled from radium and its 
 products will steadily increase until a maximum is reached 
 after 226 years. After that period, the number will decrease 
 nearly exponentially with a period of 1300 years. 
 
 The variation with time of the number of /3 particles ex- 
 pelled from radium is shown in Fig. 38, Curve BB. 
 
 Radium and its family of rapidly changing products together 
 emit four a particles for the one emitted from radium F. By 
 calculation it can be shown that the number of a particles 
 expelled from radium will reach a maximum after about 111 
 years, and will then be about 1.19 times the number emitted 
 from radium about one month old. The number, as in the case 
 of the (3 particles, will then decrease and the period will be 
 1300 years. 
 
138 
 
 RADIOACTIVE TRANSFORMATIONS 
 
 Curve AA shows the variation with time of the number of a 
 particles expelled from radium and its mixture of products. 
 Curve CC represents the number of radium atoms breaking up 
 per second. 
 
 These calculations of the variation of the activity of radium 
 with time depend upon the accuracy of the periods of change of 
 
 too 
 
 eo 
 
 n 
 
 L 
 
 200 
 
 400 
 
 600 800 
 
 TIME IN YEARS. 
 
 /OOO 
 
 /200 
 
 FIG. 38. 
 
 Curve A A represents the variation with time in the number of a particles ex- 
 pelled from radium. Curve B B represents the number of particles expelled 
 per second. Curve C C represents the number of atoms of radium breaking 
 up per second. 
 
 radium and radium D. Any alteration in these values will to 
 some extent alter the curves of variation of activity with time. 
 
 IDENTITY OF RADIUM F WITH RADIOTELLURITJM 
 
 Since the products D, E, and F are continuously produced by 
 radium, they should be found in all radioactive minerals con- 
 
SLOW TRANSFORMATIONS OF RADIUM 139 
 
 taining radium, and in amounts proportional to the amount of 
 radium in the mineral. It is now necessary to consider whether 
 any of these products have been previously separated from 
 radioactive minerals and known by other names. 
 
 We shall first consider the product radium F, which will be 
 shown to be identical with the very active substance called 
 radiotellurium, separated by Marckwald from pitchblende resi- 
 dues. In endeavoring to establish the indentity of two prod- 
 ucts, the main criteria to be relied upon are : 
 
 (1) the identity of the radiations or characteristic emanations 
 emitted by the products ; 
 
 (2) the identity of the periods of change of the products ; 
 
 (3) the similarity of chemical and ph} T sical properties of the 
 active products. 
 
 The third criterion is initially of less importance than (1) or 
 (2), since in most cases the active products are separated in a 
 very impure state and the apparent chemical reaction may be 
 largely modified by the presence of impurities. 
 
 We have seen that the product radium F emits only a rays, 
 has a period of about 143 days, and is deposited on bismuth 
 from the active solution. Radiotellurium behaves in identically 
 the same manner. In addition, the writer : directly compared 
 the rates of decay of the activities of radiotellurium and radium F 
 and found them to be the same within the limit of experimental 
 error. Each loses half its activity in about 143 days. The 
 period of decay of radiotellurium has also been experimentally 
 examined by Meyer and Schweidler and Marckwald. The former 
 found a period of 135 days and the latter 139. Considering the 
 difficulty of making accurate comparative measurements over 
 such long intervals of time, the values obtained by the different 
 observers are in remarkably good agreement. 
 
 The writer also found that the rays emitted from radium F 
 had the same penetrating power as those emitted from an active 
 bismuth plate coated with radiotellurium. It is known from 
 the work of Bragg and others, that each product of radium 
 emits a rays of a penetrating power, which is definite for each, 
 
 1 Rutherford : Phil. Mag., Sept., 1905. 
 
140 RADIOACTIVE TRANSFORMATIONS 
 
 but varies considerably among the different products. This 
 equality in penetrating power thus supplies strong evidence in 
 favor of the identity of the two products. 
 
 We may thus conclude that the radiotellurium of Marckwald 
 contains as its active constituent the product radium F; or, in 
 other words, radiotellurium is a transformation product of 
 radium. 
 
 The methods of separation and concentration of radiotel- 
 lurium used by Marckwald are of special interest. The sepa- 
 ration by Mme. Curie of radium from pitchblende in which it 
 existed in the proportion of less than one part in a million was 
 in itself a notable performance, but the work of Marckwald in 
 the separation of radiotellurium constitutes a still more strik- 
 ing illustration of the possibility of chemically concentrating a 
 radioactive substance existing in almost infinitesimal amount. 
 
 Marckwald initially observed that a bismuth rod dipped into 
 a solution of pitchblende residues became coated with a deposit 
 which emitted only a rays. After some days, the active sub- 
 stance was in this way almost completely removed from the 
 solution and obtained on the bismuth. The deposit on the 
 bismuth was found to consist for the most part of tellurium, 
 and for this reason Marckwald called the active substance radio- 
 tellurium. Later Marckwald devised very simple and efficient 
 means of separating the active matter from the tellurium, and 
 finally obtained a substance which, weight for weight, was far 
 more active than radium. 
 
 Five tons of uranium residues, corresponding to 15 tons of 
 the Joachimsthal mineral, were worked up to extract the radio- 
 tellurium from it, and he finally obtained only 3 milligrams of 
 the active substance. If plates of tin, copper, or bismuth were 
 dipped into a hydrochloric acid solution of this substance, 
 they were found to be covered with a finely divided deposit. 
 These plates were extremely radioactive, and gave marked ioniz- 
 ing, photographic, and phosphorescent effects. As an illustra- 
 tion of the enormous activity of this substance, Marckwald states 
 that a weight of 1/100 of a milligram deposited on a copper plate 
 4 sq. cms. in area lighted up a zinc sulphide screen brought 
 
SLOW TRANSFORMATIONS OF RADIUM 141 
 
 near it so strongly that the luminosity could be clearly seen by 
 an audience of several hundred people. 
 
 In consequence of the minute amount of working material, 
 Marckwald has not yet succeeded in purifying the active sub- 
 stance sufficiently to determine its spectrum. 
 
 By means of a simple calculation, the activity of radium F, 
 i.e. of radiotellurium in a pure state, can readily be deduced. 
 Let JVj be the number of atoms of radium F in one gram of 
 the radioactive mineral, and N 2 the number of radium atoms. 
 Radium and radium F are in radioactive equilibrium, and conse- 
 quently the same number of atoms of each break up per second. 
 
 Thus, X 1 N 1 = X Z N 2 , 
 
 where \, A 2 are the constants of change of radium F and 
 radium respectively. Now radium F is half transformed in 
 .38 years and radium in about 1300 years. Consequently 
 
 Now it is probable that the atomic weights of radium and 
 radium F are not very different. Consequently, for every gram 
 of radium in the mineral, there exists only .29 milligram of 
 radium F. For equal weights, the number of a particles ex- 
 pelled from radium F is 3400 times as great as the number 
 expelled from radium itself, or 850 times as great as the num- 
 ber expelled from radium about one month old, when it is in 
 equilibrium with its three rapidly changing a ray products. 
 
 Assuming that the a particle from radium F produces about 
 the same amount of ionization as the average a particle from 
 radium, the activity of radium F measured by the electric 
 method should be 850 times greater than that of radium. 
 
 It has been found experimentally that the amount of radium 
 in radioactive minerals is always proportional to their content 
 of uranium and that for every gram of uranium there is present 
 3.8 x 10~ 7 gram of radium. 
 
 The weight of radium F per gram of uranium is thus 
 1.1 x 10~ 10 gram, and per ton of 2000 Ibs., 0.1 milligram. From 
 
142 
 
 RADIOACTIVE TRANSFORMATIONS 
 
 15 tons of Joachimsthal mineral, which contains about 50 per cent 
 of uranium, the yield of radium F should be 0.75 milligrams. 
 
 The amount separated by Marckwald from this amount of 
 pitchblende was about 3 milligrams. It is unlikely that the 
 whole amount of radium F was separated, and the three milli- 
 grams probably contain some impurity. The theoretical pro- 
 portion of radium F in the radioactive mineral is thus in good 
 agreement with the experimental results. 
 
 Although the proportion of radium F in minerals may appear 
 extremely small, yet the other more rapidly changing products 
 exist in still smaller amounts. The weight of each product 
 present per ton of uranium is directly proportional to its period, 
 so that the most swiftly changing product is present in the 
 smallest quantity. In the following table are shown the weights 
 of each of the radium products present per ton of 2000 pounds 
 of uranium in the mineral. 
 
 Products. 
 
 Period. 
 
 Weight in milligrams per ton 
 of uranium. 
 
 Radium 
 
 1300 years 
 
 340 milligrams 
 
 Emanation 
 Radium A 
 
 3.8 days 
 3 minutes 
 
 2.6 X lO- 3 milligrams 
 1.4 X 10- 6 
 
 B 
 
 26 
 
 1.2 X 10-* 
 
 C 
 
 19 
 
 9 X 10- 6 
 
 " D 
 
 40 years 
 
 10 
 
 E 
 
 6 days 
 
 4.2 X 10-3 
 
 F 
 
 143 " 
 
 .1 
 
 The products radium A, B, C, and E exist in far too small 
 amounts to be examined by ordinary chemical methods, even if 
 their short life allowed it. Radium D, however, is present in 
 considerable quantity compared with radium F, and it should 
 be possible to obtain a sufficient amount of it for a chemical 
 examination. 
 
 POLONIUM AND RADIOTELLUKIUM 
 
 It will be remembered that the first active substance separated 
 from pitchblende was found associated with bismuth and was 
 called polonium by its discoverer, Mme. Curie. 
 
SLOW TRANSFORMATIONS OF RADIUM 143 
 
 Several methods were devised for the concentration of the 
 active material mixed with the bismuth, and Mme. Curie finally 
 succeeded in obtaining an active substance comparable in activ- 
 ity with radium. The polonium gave out only a rays, and its 
 activity was not permanent but gradually decreased. 
 
 Both as regards the nature of its rays and its physical and 
 chemical properties, polonium is very analogous to the product 
 radium F and radiotellurium. There has been a considerable 
 amount of discussion at various times as to whether the active 
 constituent in radiotellurium is identical with that present 
 in polonium. At first it was announced that the activity of 
 radiotellurium did not decay appreciably, and it apparently 
 behaved in this respect quite differently from polonium. We 
 now know that radiotellurium does lose its activity, and fairly 
 rapidly. 
 
 If the two products contain the same constituent, their ac- 
 tivities should decay according to the same period. Mme. 
 Curie, however, has observed that some of her preparations of 
 polonium do not lose their activity according to an exponential 
 law. 
 
 For example, a sample of polonium nitrate lost half its 
 activity in 11 months and 95 per cent in 33 months. A sample 
 of the metal lost 67 per cent of its activity in 6 months. These 
 results are not at all concordant. The sample of the metal 
 loses its activity slightly faster than radium F, while the nitrate 
 at first loses it much more slowly. If these results are reliable, 
 the divergence of the activity curves from the exponential law 
 shows that more than one substance is present in the polonium 
 experimented with by Mme. Curie. It is very probable that 
 this second constituent is radium D. The presence of some of 
 this substance, which gives rise to radium F, would cause the 
 a ray activity to decrease at first more slowly than the normal 
 rate when only radium F is present. 
 
 Considering the similarity of polonium and radiotellurium in 
 their chemical, physical, and radioactive properties and the 
 probable identity of their periods, I think that there can be no 
 doubt that the a ray constituent in polonium is the same as 
 
144 RADIOACTIVE TRANSFORMATIONS 
 
 that separated by a different method by Marckwald. We may 
 then conclude that radiotellurium and polonium are both de- 
 rived from the transformation of the radium atom. 1 
 
 CONNECTION OF RADIOLEAD WITH THE ACTIVE DEPOSIT 
 
 We shall now describe some experiments which show con- 
 clusively that the product radium D is the primary constituent 
 of the radiolead, first separated by Hofmann from pitchblende 
 residues. The early results of Hofmann, on the separation and 
 properties of radiolead, were subjected to considerable criticism, 
 but there is now no doubt that to him belongs the credit of 
 separation of a new product from pitchblende, which proves to 
 be the parent of radiotellurium and polonium, 
 
 My attention was first drawn to the connection between 
 radiolead and the active deposit of radium by an examination 
 of a specimen of radiolead kindly prepared for me by Dr. Bolt- 
 wood of New Haven. This was found to give out initially an 
 unusual proportion of ft rays as compared with its a ray activity, 
 and the a ray activity was found to increase progressively with 
 time. In these respects it behaved in a similar manner to a 
 substance initially containing radium D and E, in which the 
 product radium F was being gradually formed, thus giving rise 
 to the increasing a ray activity. 
 
 The connection of radiolead with radium D, E, and F has 
 been conclusively proved by a chemical examination of the 
 radioactive constituents found in radiolead and a determination 
 of their periods of decay. It must be borne in mind that the 
 name radiolead was given to the active substance because it 
 was first separated mixed with lead, but we now know that 
 the active substances contained in it have no more connection 
 with lead than radium has with the barium from which it is' 
 finally separated. 
 
 1 The identity of the active constituent of radiotellurium and polonium has now 
 been definitely settled. Mme. Curie (Comptes rendus, Jan. 29, 1906) has accurately 
 determined the loss of activity of polonium, and found that it decayed according to 
 an exponential law with a period of 140 days. This period of decay is practically 
 identical with that found for radiotellurium and radium F. 
 
SLOW TRANSFORMATIONS OF RADIUM 145 
 
 Hofmann, Gonders, and Wolfl. 1 in the course of a chemical 
 examination of a specimen of radiolead obtained the following 
 results. Experiments were first made on the effect of adding 
 substances to a solution of radiolead and then removing them 
 by precipitation. Small quantities of the platinum metals in 
 the form of chlorides were left in the solution for several weeks 
 and then precipitated by formalin or hydroxylamine. All of 
 these substances after separation were found to give out a and 
 ft rays. 
 
 Most of the ft ray activity disappeared in about six weeks, and 
 the a ray activity in about one year. We shall see that the ft ray 
 activity is due to the separation of radium E, which decays to 
 half value in 6 days, while the a ray activity is due to radium F. 
 This conclusion is further confirmed by experiments on the 
 effect of heat on the activity of these substances. At a full red 
 heat, the a ray activity was lost in a few seconds. This is in 
 agreement with experiments on radium F which is volatilized 
 at about 1000 C. 
 
 Salts of gold, silver, and mercury, added to the solution of 
 radiolead, were found to show only a ray activity. This is 
 explained if radium F is alone removed. Bismuth salts, on the 
 other hand, showed a and ft activity, but the latter died away 
 rapidly. This shows that bismuth removes both radium E 
 and F. 
 
 The a ray activity of the radiolead is much reduced by pre- 
 cipitation of bismuth in the solution, but gradually increases 
 again with time. This result is exactly what is to be expected 
 if radiolead contains radium D, E, and F. Radium E and F 
 are removed with the bismuth, while D is left behind, and in 
 consequence there is a fresh growth of radium E and F. 
 
 The radioactive measurements made by Hofmann, Gonders, 
 and Wolfl were unfortunately not very precise, but this want 
 has been supplied by some recent careful measurements by 
 Meyer and Schweidler. 2 If radiolead contains radium D, E, 
 and F, the ft ray activity due to E should decay to half value 
 
 1 Hofmann, Gonders, and Wolfl: Ann. d. Phys ., v, p. 615 (1904). 
 
 2 Meyer and Schweidler: Wien Ber., July, 1905. 
 
 10 
 
146 RADIOACTIVE TRANSFORMATIONS 
 
 in 6 days, and the a ray activity due to F to half value in 
 about 140 days. 
 
 These results have been completely confirmed by Meyer and 
 Schvveidler, who have accurately measured the rates of decay 
 of the various products from radiolead. A series of palladium 
 plates were immersed in the radiolead solution. After removal 
 the activity consisted of a and ft rays. The /3 ray activity 
 decreased exponentially with the time, falling to half value in 
 6.2 days. The ft ray product is thus identical with radium E. 
 The a ray activity, after some months, fell off exponentially 
 with a period of 135 days. The a ray product is thus identical 
 with radium F. 
 
 There is thus no doubt that radiolead some time after its 
 preparation contains radium D, E, and F. No observations 
 have so far been made to settle definitely whether radium D, E, 
 and F are removed together with the lead or whether only ra- 
 dium D is removed and the presence of radium E and F after 
 some time is due to their production from radium D. If the 
 bismuth is separated from the lead, it seems likely that radium E 
 and F would be removed with the former and that radium D 
 alone would be removed with the lead. 
 
 It is thus seen that the primary constituent in radiolead is 
 the parent of radiotellurium and polonium. 
 
 The connection of the radium products with radiolead is out- 
 lined in the following table. 
 
 Radium D = product in new radiolead. No rays. Half transformed 
 in 40 years. 
 
 Radium E gives out ft rays ; is separated with bismuth, iridium, 
 and palladium. Half transformed in 6 days. 
 
 Radium F = product in polonium and radiotellurium. Gives out 
 only a rays. Volatile at 1000 C., and attaches it- 
 self to bismuth and palladium. Half transformed 
 in 143 days. 
 
 These results have thus emphasized the importance of radium 
 D as a new radioactive substance in pitchblende. 
 
SLOW TRANSFORMATIONS OF RADIUM 147 
 
 It has been shown that about 10 milligrams of radium D 
 should be separated from the mineral for each ton of uranium 
 present. A few weeks after separation the /8 ray activity of 
 this substance should be about 30 times as great as that of 
 radium. A quantity of this substance would serve as a useful 
 source of /3 rays, and also as a very convenient means of obtain- 
 ing radium F. A very active deposit of this substance could 
 at any time be obtained by placing a bismuth or palladium plate 
 in the solution. It is to be hoped that this substance will be 
 separated from pitchblende residues at the same time as radium, 
 for in many respects it would prove as useful in experiments as 
 radium itself. 
 
CHAPTER VI 
 ORIGIN AND LIFE OF RADIUM 
 
 Since radium itself continuously throws off a particles and 
 gives rise to a radioactive gas, its amount must steadily de- 
 crease with time. Radium, in this respect, must be considered 
 as a radioactive product like the emanation, the only difference 
 being its comparatively slow rate of change. A given amount 
 of radium left to itself must ultimately disappear as such, and 
 after a series of transformations there will only remain the 
 inactive substances produced by its decomposition. 
 
 The time of observation has been far too short to fix the 
 period of change of radium by direct experiment. Accurate 
 measurements of the activity will not supply any information 
 of value on this point for a long interval, since the slow trans- 
 formation products arising from the radium actually cause a 
 steady increase of activity for several hundred years. 
 
 There are several indirect methods which can be employed 
 to deduce the probable period of radium, depending on (1) the 
 number of a particles expelled per second, (2) the observed 
 heating effect of radium, and (3) the observed volume of the 
 emanation released from it. 
 
 Method 1. We shall first consider the method based on the 
 rate of expulsion of a particles. By measuring the charge 
 carried by the a rays expelled from a thin film of radium, the 
 writer 1 found that the total number of a particles expelled per 
 second from one gram of radium at its minimum activity was 
 6.2 x 10 10 , assuming that each a particle carries with it the 
 usual ionic charge of 3.4 xlO~ 10 electrostatic units. When in 
 radioactive equilibrium with its family of rapidly changing 
 
 1 Rutherford: Phil. Mag., Aug., 1905. 
 
ORIGIN AND LIFE OF RADIUM 149 
 
 products, the number of expelled particles is four times as. 
 great. 
 
 The simplest assumption is that one a particle is expelled 
 from each atom as it breaks up. Consequently 6.2 x 10 1(> 
 atoms of radium break up per second. Now it has been found 
 from data based upon experiment that one cubic centimetre of a 
 gas, hydrogen, for example, at standard pressure and tem- 
 perature contains 3.6 x 10 19 molecules. From this it follows 
 that one gram of radium of atomic weight 225 contains 
 3.6xl0 21 atoms of radium. The fraction of radium which 
 breaks up per second is 
 
 6.2x10" 
 3.6 X 10 21 - 
 
 or 5.4 x lO" 4 per year. 
 
 Like any other active product, the amount of radium must 
 decrease according to an exponential law, so that the value of 
 its constant of change \ is 5.4 x 10" 4 (year)" 1 . 
 
 From this it follows that half of the radium is transformed in 
 about 1300 years. The average life of the radium atom which 
 is measured by I/A, is about 1800 years. 
 
 Method 2. The calculation of the life of radium can also be 
 based on the observed heating effect of radium, which will be 
 shown later (Chapter X) to be a direct measure of the kinetic 
 energy of the expelled a particles. From measurements of the 
 velocity and mass of the a particle expelled from radium, the 
 average energy of motion, mv 2 , of the a particle was found by 
 the writer 'to be 5.9 x 10" 6 ergs. Now it is found experi- 
 mentally that one gram of radium emits heat at the rate of 
 about 100 gram calories per hour. If this is due to the kinetic 
 energy of the a particles, the number of such particles that 
 must be expelled per second is about 2.0 x 10 n . The number 
 from radium itself is one quarter of this. Using the same 
 method of calculation as before, it is seen that half of the 
 radium is transformed in about 1600 years a value not very 
 different from that deduced by the first method. 
 
 Method 3. We shall now consider the calculation of the 
 
150 RADIOACTIVE TRANSFORMATIONS 
 
 life of radium based on the observed volume of the emanation 
 released from one gram of radium. Ramsay and Soddy found 
 that this maximum volume was slightly greater than one cubic 
 millimetre at standard pressure and temperature. Now one 
 cubic millimetre of gas contains 3.6 x 10 19 molecules. The 
 number of molecules of emanation produced per second is X 
 times the equilibrium number present, where X is the constant 
 of change of the emanation. Assuming, as is probably the 
 case, that the emanation is a monatomic gas, and that each 
 atom of radium in breaking up gives rise to one atom of emana- 
 tion, the number of atoms of radium breaking up per second is 
 7.6 x 10 10 . Proceeding as before, this gives 1050 years as the 
 period of radium. 
 
 The first two methods involve the assumption of the number 
 of atoms present in one cubic centimetre of a gas. The calcu- 
 lation based on the volume of the emanation can, however, be 
 made in a different way without this assumption. If one atom 
 of radium by the loss of one a particle is changed into one atom 
 of emanation, the molecular weight of the latter must be at 
 least 200. The value deduced from experiments on diffusion 
 is about 100, but on page 85 some reasons have been given in 
 support of the view that this is an underestimate. One cubic 
 millimetre of the emanation thus weighs as much as 100 c.mms. 
 of hydrogen, i. e. 8.96 x 10" 6 grams. The weight of emanation 
 produced per second is X times this amount, i. e. 1.9 x 10~ u 
 grams. The weight of emanation produced per year is thus 
 6 x 10" 4 grams, and this must be nearly equal to the weight of 
 radium breaking up per year. This makes the period of radium 
 about 1300 years. 
 
 Considering the uncertainty attaching to the exact values of 
 the data used in these calculations, the periods deduced by the 
 three methods are in good agreement. In calculations we shall 
 take 1300 years as the most probable value of the period of 
 radium. 
 
 Radium thus breaks up at a fairly rapid rate, and in the 
 course of a few thousand years a mass of radium left by itself 
 
ORIGIN AND LIFE OF RADIUM 151 
 
 would lose a large proportion of its activity. Assuming that 
 radium breaks up with a period of 1300 years, it can readily be 
 calculated that after an interval of 26,000 years, only one mil- 
 lionth of the mass of radium would remain unchanged. If we 
 suppose, for illustration, that the earth was originally composed 
 of pure radium, the activity observed in the earth 26,000 years 
 later would be about the same as that observed to-day in a good 
 specimen of pitchblende. This period of years is very small 
 compared with the age of the minerals of the earth, and unless 
 the very improbable assumption is made, that the radium was 
 in some way suddenly formed at a very late period in the 
 earth's history, we are forced to the conclusion that radium 
 must be continuously produced in the earth. It was early sug- 
 gested by Rutherford and Soddy that radium might be a disinte- 
 gration product of one of the radioactive elements present in 
 pitchblende. Both uranium and thorium fulfil the conditions 
 required as a possible parent of radium. Both have atomic 
 weights greater than that of radium, and both are transformed at 
 a very slow rate compared with radium. A cursory examination 
 shows that uranium is the most likely parent, since radium* is 
 always found in largest amount in uranium minerals, while some 
 thorium minerals contain very little radium. 
 
 We shall now consider some of the consequences that should 
 follow if uranium is considered to be the parent of radium. 
 
 Several thousand years after the uranium has been formed, the 
 amount of radium should reach a definite maximum value. Its 
 rate of production by the uranium is then balanced by its own 
 rate of disappearance. Under such conditions, the number of 
 atoms of radium which break up per second is equal to the 
 number of atoms of uranium which break up per second. Now, 
 as far as observation has gone, uranium only emits one a par- 
 ticle during its transformation into uranium X. The product 
 uranium X does not emit a rays, but only /3 and 7 rays. On 
 the other hand, we have seen that radium itself and four of its 
 products, viz., the emanation, radium A, C, and F, emit a rays. 
 The number of a particles expelled from the radium, and these 
 products of its transformation, should thus be five times the 
 
152 RADIOACTIVE TRANSFORMATIONS 
 
 number expelled from uranium. Assuming that the a particles 
 from the radium products produce about the same ionization as 
 the a particle from uranium, the activity of a radioactive mineral, 
 which consists mostly of uranium, should be about six times that 
 of uranium itself. Now the best pitchblende shows an activ- 
 ity about five times that of uranium, so that the theoretical 
 result is approximately realized in practice. Until, however, the 
 relative ionizations produced by the a particles from uranium 
 and each of the radium products are accurately known, the 
 relative activities to be expected cannot be fixed with certainty. 
 
 Another consequence of the theory is that the amount of 
 radium in any radioactive mineral should be always proportional 
 to its content of uranium. This must hold in every case, pro- 
 vided neither the uranium nor the radium have been removed 
 from the mineral by physical or chemical action. This interest- 
 ing question has been experimentally attacked by Boltwood, 1 
 Strutt, 2 and McCoy, 3 and has yielded results of the highest 
 importance. 
 
 McCoy accurately compared the activities of different radio- 
 active minerals, and showed that in every case the activity was 
 very nearly proportional to their percentage content of uranium. 
 Since, howejfc", the radioactive minerals contain some actinium, 
 and occasionally some thorium, these results indicate that the 
 activity of all these substances, taken together, is proportional 
 to the amount of uranium. Boltwood and Strutt employed a 
 more direct method, by determining the relative content of 
 uranium and radium in radioactive minerals. The amount of 
 uranium was determined by direct chemical analysis, while the 
 amount of radium was determined by measurements of the 
 amount of radium emanation released by solution of the mineral. 
 The relative amount of the latter can be determined with great 
 accuracy by the electric method, which is the most convenient 
 method of comparing quantitatively the amounts of radium in 
 different minerals. 
 
 1 Boltwood: Nature, May 25, 1904; Phil. Mag., April, 1905. 
 
 2 Strutt: Trans. Roy. Soc. A., 1905. 
 
 8 McCoy : Ber. d. d. chem. Ges., No. 11, p. 2641, 1904. 
 
ORIGIN AND LIFE OF RADIUM 
 
 153 
 
 The results of both these observers show that there is a nearly 
 constant ratio between the amounts of radium and uranium in 
 every mineral examined, except in one case, which will be con- 
 sidered later. Minerals were obtained from various localities, 
 both in Europe and America, which varied widely in chemical 
 composition and in the percentage content of uranium. The 
 experiments of Dr. Boltwood of Yale University, which have 
 been made with great care and accuracy, show a surprisingly 
 constant ratio between the amounts of uranium and radium. 
 
 A brief account will be given of the methods employed by 
 him in his measure- 
 ments. The percentage 
 of uranium in the min- 
 eral under consideration 
 was first determined by 
 chemical analysis. A 
 known weight of the 
 finely powdered mineral 
 was placed in a glass 
 vessel A (Fig. 39), and 
 sufficient acid intro- 
 duced to dissolve it. 
 
 The acid was then 
 boiled until the mineral 
 was completely dis- 
 solved, and the emana- 
 tion mixed with air was collected on the top of the column of 
 water in the tube D. This emanation was then introduced into 
 a closed electroscope of the type shown in Fig. 6, page 29, which 
 was first exhausted. Air was then introduced until the gas 
 inside the electroscope was at atmospheric pressure. On ac- 
 count of the excited activity produced by the emanation, the 
 rate of discharge of the electroscope did not reach a maximum 
 until about three hours after the introduction of the emana- 
 tion. The rate of movement of the gold leaf of the elec- 
 troscope was taken as a measure of the amount of emanation 
 present. The emanations of thorium or actinium, released 
 
 FIG. 
 
154 RADIOACTIVE TRANSFORMATIONS 
 
 from the mineral at the same time as the radium emanation, 
 had, on account of the rapid decay of their activity, completely 
 disappeared before the introduction of the radium emanation 
 into the electroscope. This process was repeated for all the 
 minerals examined. 
 
 Boltwood observed that some minerals had considerable eman- 
 ating power, i. e., the minerals lost some of their emanation 
 when in the solid state. Under these conditions, the amount 
 of emanation released by solution and boiling of the mineral 
 would be less than the equilibrium amount. The proper correc- 
 tion was made by sealing up a known weight of the mineral in a 
 tube for one month and then measuring with the same electro- 
 scope the amount of emanation which collected in the air above 
 the mineral. The sum of the two amounts gives the true 
 equilibrium quantity of emanation corresponding to the radium 
 present in the mineral. 
 
 The results obtained by Boltwood are shown in the following 
 table. The numbers in column I give, in arbitrary units, the 
 amount of emanation released by solution and boiling; column II 
 shows the percentage of the emanation which escaped into the 
 air ; column III shows the amount of uranium in the mineral ; 
 and column Jjt the numbers obtained by dividing the equilib- 
 rium amoun^m emanation by the quantity of uranium present. 
 
 If the amount of radium always bears a definite ratio to the 
 amount of uranium, the numbers in column IV should be the 
 same. With the exception of some of the monazites, there is 
 a remarkably good agreement, and, taking into consideration 
 the great variation in the amount of uranium in the different 
 minerals, and the wide range of locality from which they were 
 obtained, the results afford a direct and satisfactory proof that 
 the amount of radium in minerals is directly proportional to the 
 amount of uranium present. 
 
 As an example of the confidence to be placed in this ratio as a 
 physical constant for all radioactive minerals, Boltwood observed 
 that some of the monazites contained a considerable quantity of 
 radium, although the previous analyses had not shown any 
 uranium to be present. A careful examination was made to 
 
ORIGIN AND LIFE OF RADIUM 
 
 155 
 
 Substance. 
 
 Locality. 
 
 I. 
 
 II. 
 
 in. 
 
 IV. 
 
 Uraninite 
 
 North Carolina 
 
 170.0 
 
 11.3 
 
 0.7465 
 
 228 
 
 Uraninite 
 
 Colorado 
 
 155.1 
 
 5.2 
 
 06961 
 
 223 
 
 Gummite 
 
 North Carolina 
 
 147.0 
 
 13.7 
 
 0.6538 
 
 225 
 
 Uraninite 
 
 Joachimsthal 
 
 139.6 
 
 5.6 
 
 0.6174 
 
 226 
 
 Uranophane 
 
 North Carolina 
 
 117.7 
 
 8.2 
 
 0.5168 
 
 228 
 
 Uraninite 
 
 Saxony 
 
 115.6 
 
 2.7 
 
 0.5064 
 
 228 
 
 Uranophane 
 
 North Carolina 
 
 113.5 
 
 22.8 
 
 0.4984 
 
 228 
 
 Thorogummite 
 
 North Carolina 
 
 72.9 
 
 16.2 
 
 0.3317 
 
 220 
 
 Carnotite 
 
 Colorado 
 
 49.7 
 
 16.3 
 
 0.2261 
 
 220 
 
 Uranothorite 
 
 Norway 
 
 25.2 
 
 1.3 
 
 0.1138 
 
 221 
 
 Samarskite 
 
 North Carolina 
 
 23.4 
 
 0.7 
 
 0.1044 
 
 224 
 
 Orangite 
 
 Norway 
 
 23.1 
 
 1.1 
 
 0.1034 
 
 223 
 
 Euxinite 
 
 Norway 
 
 19.9 
 
 0.5 
 
 0.0871 
 
 228 
 
 Thorite 
 
 Norway 
 
 16.6 
 
 6.2 
 
 0.0754 
 
 220 
 
 Fergusonite 
 
 Norway 
 
 12.0 
 
 0.5 
 
 0.0557 
 
 215 
 
 Aeschynite 
 Xenotine 
 
 Norway 
 Norway 
 
 10.0 
 1.54 
 
 0.2 
 
 26.0 
 
 0.0452 
 0.0070 
 
 221 
 220 
 
 Monazite (sand) 
 
 North Carolina 
 
 0.88 
 
 
 0.0043 
 
 205 
 
 Monazite (crys.) 
 
 Norway 
 
 0.84 
 
 1.2 
 
 0.0041 
 
 207 
 
 Monazite (sand) 
 
 Brazil 
 
 0.76 
 
 . 
 
 0.0031 
 
 245 
 
 Monazite (massive) 
 
 Connecticut 
 
 0.63 
 
 
 0.0030 
 
 210 
 
 test this point, and it was found that uranium was present in 
 the amount to be expected according to theory. The failure to 
 detect the presence of uranium in the earlier analysis was due 
 to the presence of phosphates. 
 
 There is one interesting apparent exception to this constancy 
 of the ratio between the amounts of uranium and radium. 
 Danne recently found that considerable quantities of radium 
 are present in certain deposits in the neighborhood of Issy 
 1'Eveque in the Saone-et-Loire district, but that no trace of 
 uranium could be detected. The active matter is found in 
 pyromorphite (phosphate of lead), and in clays containing 
 lead, but the radium is usually found in greater quantities in 
 the former. The pyromorphite is found in veins of quartz and 
 felspar rocks. The veins were always wet, owing to the pres- 
 ence of springs in the neighborhood. The percentage of uranium 
 in the pyromorphite varies considerably for different specimens, 
 but Danne states that on an average one centigram of radium 
 is present per ton. 
 
 It seems probable that this radium has been deposited in the 
 rocks after being carried from a distance by means of under- 
 
156 RADIOACTIVE TRANSFORMATIONS 
 
 ground springs. The presence of radium in this district is not 
 surprising, for crystals of autunite have been found about forty 
 miles distant. This result is of interest, for it suggests that 
 radium can in some cases be removed from the radioactive 
 mineral by solution in water, and be deposited under suitable 
 physical and chemical conditions some distance away. It also 
 suggests the possibility that deposits containing a considerable 
 proportion of radium may yet be discovered in positions where 
 the conditions necessary for the solution and re-deposit of the 
 radium are favorable. 
 
 AMOUNT OF RADIUM IN MINERALS 
 
 The weight of radium in a mineral, per gram of uranium, is 
 thus a definite constant of considerable practical as well as 
 theoretical importance. This constant was recently determined 
 by Rutherford and Boltwood by comparison of the amount of 
 emanation liberated from a known weight of uraninite with 
 that released from a known quantity of pure radium bromide 
 in solution. For the latter purpose, a known weight of radium 
 bromide was taken from a sample of radium bromide obtained 
 from the Quinin Fabrik, Braunschweig, which had previously 
 been found to give out heat at a rate of over 100 gram calories 
 per hour. P. Curie and Laborde found that their pure radium 
 chloride preparations gave out heat at the rate of about 100 gram 
 calories per hour. We may thus conclude that the radium prep- 
 aration employed was nearly pure. This known weight about 
 one milligram was dissolved in water, and by successive dilu- 
 tions a standard solution was made up containing 10" 6 grams of 
 radium bromide per cubic centimetre. Taking the constitution 
 of radium bromide as RaBr 2 and the atomic weight of radium as 
 225, it was deduced that in each gram of uranium in the mineral 
 the corresponding weight of radium was 3.8 x 10~ 7 gram. 1 
 
 From this it follows that .34 gram of radium is present in 
 a mineral per ton of uranium. Since the radioactive minerals 
 
 1 The first determination of this constant by Rutherford and Boltwood (Amer. 
 Journ. Sci., July, 1905) gave a value of 7.4 X 10~ 7 . This was later found by them 
 to be incorrect, owing to a precipitation of the radium in the standard solution. 
 
ORIGIN AND LIFE OF RADIUM 157 
 
 from which radium is extracted usually contain about 50 per 
 cent of uranium, the yield of radium per ton of mineral should 
 be about .17 gram. 
 
 Assuming, as a first approximation, that the a particles from 
 radium and its products, and from uranium, are expelled at the 
 same speed, the activity of the radium and its family of rapidly 
 changing products when in equilibrium with the uranium, 
 should be four times that of uranium. Taking the activity of 
 pure radium as about three million times that of uranium, the 
 weight of radium required to produce this activity is 
 
 A 
 
 1.33 X 10- 6 grams. 
 
 3 x 10 6 
 
 The observed amount, 3.8 x 10~ 7 grams, is considerably 
 smaller. The agreement between theory and experiment, how- 
 ever, becomes much closer when we take into account the known 
 fact that the average a particle from radium has a greater pene- 
 trating power and consequently produces a greater number of ions 
 in the gas than the average a particle expelled from uranium. 
 
 GROWTH OF RADIUM IN URANIUM SOLUTIONS 
 
 Although the constancy of the ratio between the amounts of 
 radium and uranium in all radioactive minerals, as well as the 
 agreement between the theoretical and observed quantities, afford 
 very strong proof of the truth of the theory that uranium is the 
 parent of radium, yet this conclusion cannot be considered as 
 completely established until it has been experimentally shown 
 that radium gradually collects in uranium solutions originally 
 freed from it. 
 
 The rate of production of radium on the disintegration theory 
 can readily be estimated. The fraction of radium breaking up 
 per year has been calculated on page 149 and shown to be about 
 5.4 x ID" 4 per year. The amount of radium per gram of 
 uranium in minerals has been shown to be 3.8 x 10~ 7 grams. 
 Consequently, in order to keep up the quantity of radium in a 
 mineral at a constant amount, the rate of supply per year per 
 gram of uranium must be 5.4 x 1Q- 4 x 3.8 x 10~ 7 = 2 x 1Q- 10 
 
158 RADIOACTIVE TRANSFORMATIONS 
 
 gram. This represents the amount of radium formed per year 
 from each gram of uranium. The presence of radium can readily 
 be detected by its emanation. Using a kilogram of uranium, 
 the amount of radium formed per year is 2 x 10~ 7 gram. The 
 emanation from this would cause a gold-leaf electroscope to be 
 discharged in a few seconds, while the amount of radium pro- 
 duced in a single day should be easily measurable. 
 
 Experiments on the growth of radium in uranium were first 
 undertaken by Soddy. 1 A kilogram of uranium nitrate in 
 solution was employed. This was first chemically treated, to 
 remove most of the small quantity of radium originally present, 
 and was then allowed to stand in a closed vessel. The equilib- 
 rium amount of emanation formed in the solution was then 
 tested at intervals. Preliminary experiments showed that the 
 rate of production of the radium was certainly far slower than 
 the theoretical value, and at first little if any indication of pro- 
 duction of radium was observed. In later experiments, how- 
 ever, Soddy found that in the course of eighteen months, the 
 amount of radium in the solution had distinctly increased. 
 
 The solution after this interval contained about 1.6 x 10~ 9 
 gram of radium. This gives the value of about 2 x 10~ 12 as 
 the fraction of uranium changing per year, while the theoreti- 
 cal fraction is 2 x 10~ 10 , or 100 times greater than the observed 
 amount. 
 
 Whetham also found a similar result, but concluded that the 
 rate of production was faster than that observed by Soddy. On 
 the other hand, Boltwood finds no certain evidence of the growth 
 of radium from uranium, although an extremely minute quan- 
 tity was detectable in his apparatus. In his experiments, 100 
 grams of uranium were obtained almost completely free from 
 radium by fractional crystallization. After this treatment, no 
 trace of radium could be detected in his uranium solution, 
 although he could with certainty have detected the presence of 
 1.7 X 10" 11 grams. 
 
 After standing for a year, no effect was produced by the 
 emanation in his electroscope, which was of the same sensitive- 
 
 1 Soddy: Nature, May 12, 1904, Jan. 19, 1905; Phil. Mag., June, 1905. 
 
ORIGIN AND LIFE OF RADIUM 
 
 ness as in the first experiments. Such a result shows that 
 uranium, when purified in the manner adopted by Boltwood, 
 certainly does not, in the course of a year, grow a measurable 
 quantity of .radium, and that the quantity is not more than one 
 thousandth of the theoretical amount. 
 
 Although the experimental evidence is somewhat conflicting, 
 I think there can be little doubt that the uranium of Soddy did 
 show a growth of radium, although only a fraction of the 
 amount to be expected theoretically. So far as is known at 
 present, uranium breaks up with the expulsion of an a particle 
 and produces uranium X, which has a period of 22 days and 
 emits only ft and 7 rays. No further active product has been 
 detected, so that we are unable to say what further stages of 
 disintegration appear before radium is formed. If, for example, 
 the disintegration product of UrX is a rayless substance with 
 a very slow period, the slow rate of production of radium by 
 uranium is at once explained. Suppose, for -example, that the 
 uranium, as in the experiments of Boltwood, was carefully puri- 
 fied. It is probable that the rayless product would be com- 
 pletely removed from the uranium. Before radium could be 
 produced at an appreciable rate, the intermediate rayless prod- 
 uct must be formed in some quantity. If the rayless product 
 had a period of several thousand years, an interval of several 
 years would be required before the appearance of radium could 
 be detected. 
 
 Such an hypothesis of an intermediate transition product 
 would also account for the discrepancy between the experi- 
 ments of Soddy and Boltwood. In the experiments of the 
 former, the trace of radium initially observed in the uranium 
 was partly removed by the precipitation of barium in the 
 uranium solution. This may not have removed the inter- 
 mediate product which had been collecting in the uranium for 
 several years. Consequently, the unpurified solution used by 
 Soddy was better suited to show the production of radium than 
 the carefully treated solution used by Boltwood. 
 
 I think that there can be no reasonable doubt that the pure 
 uranium solution will ultimately show the presence of radium, 
 
160 RADIOACTIVE TRANSFORMATIONS 
 
 although an interval^ of several years may be required before the 
 amount formed is detectable. 
 
 The changes occurring in uranium which lead to the pro- 
 duction of radium are shown below. 
 
 Uranium. 
 
 Uranium X. 
 
 1 
 
 One or more unknown transition substances with long 
 
 periods of transformation. 
 
 f 
 
 Kadiuin and its family of products. 
 
 There can be little doubt that the intermediary product or 
 products between uranium X and radium will ultimately be sepa- 
 rated chemically. Supposing that there is only one intermediary 
 product, it is not unlikely that this will prove to be rayless in 
 character. The presence of such a product could be detected 
 by its property of producing radium initially at a constant rate. 
 If, for example, the unknown product were completely separated 
 from an amount of radioactive mineral which contained a kilo- 
 gram of uranium, it would produce radium initially at the rate 
 of about 4 x 10~ 7 gram per year, or 10~ 9 of a gram per day. 
 This latter amount is easily measurable, and consequently a 
 proof of the production of radium by this substance should only 
 require observations extending over a few weeks. 
 
 The position that radium holds in regard to uranium is unique 
 in chemistry. For the first time it is possible to predict accu- 
 rately the amount of one element present when the quantity of 
 another is known. In seems probable that such relations will 
 ultimately be extended to include all the radioelements and 
 their products, and possibly also some of the apparently non- 
 radioactive substances; for it is remarkable how certain ele- 
 ments are always found together in mineral deposits in about 
 the same relative amounts, although there is no apparent chem- 
 ical reason for their association. 
 
CHAPTER VII 
 
 TRANSFORMATION PRODUCTS OF URANIUM AND ACTINIUM, 
 AND THE CONNECTION BETWEEN THE RADIOELEMENTS 
 
 We have in previous chapters analyzed in some detail the 
 series of transformations that take place in thorium and radium. 
 As the two other radioactive substances, uranium and actinium, 
 are also of interest in this connection, a brief review will now 
 be given of the changes taking place in them. 
 
 CHANGES IN URANIUM 
 
 Uranium products give out a, /3, and 7 rays, but no definite 
 evidence has yet been obtained that uranium gives off an ema- 
 nation. In this respect, it appears to differ from thorium, 
 radium, and actinium. It is, however, possible that a closer 
 investigation may yet disclose the presence of an emanation 
 with a very short life. If an emanation were emitted which 
 lasted for less than a hundredth part of a second, its detection 
 by the electric method would be extremely difficult. 
 
 Only one direct transformation product, called uranium X, has 
 so far been observed in uranium. The separation of this sub- 
 stance was first effected by Sir William Crookes 1 by two dis- 
 tinct methods. Ammonium carbonate in excess was added to 
 a uranium solution and the uranium precipitated. A light 
 precipitate remained behind which contained the UrX. Crookes 
 used the photographic method and observed that the uranium, 
 after this treatment, was photographically almost inactive, 
 while the precipitate containing the UrX, when compared with 
 an equal weight of uranium, had a very intense photographic 
 action. The explanation of this was made clear by later ex- 
 periments. UrX gives out only /3 rays, which, in the case of 
 uranium, produce far more photographic action than the easily 
 
 1 Crookes: Proc. Roy. Soc., Ixvi, p. 409 (1900). 
 11 
 
162 RADIOACTIVE TRANSFORMATIONS 
 
 absorbed a rays. The removal of UrX does not in any way 
 alter the a ray activity of uranium, measured by the electric 
 method, but completely removes the /3 ray activity. 
 
 The second method used by Crookes was to dissolve uranium 
 in ether, when the uranium divides itself unequally between 
 the ether and water present. The water fraction contains all 
 the UrX, while the ether fraction is photographically inactive. 
 
 Still another means of separation of UrX was used by Bec- 
 querel. 1 A small quantity of a barium salt was added to a 
 uranium solution, and then precipitated by the addition of sul- 
 phuric acid. The dense barium precipitate carries down the 
 UrX with it, and, after several successive treatments, the UrX 
 is almost completely removed from the uranium. Becquerel 
 first noted that the UrX loses its activity after some time, while 
 the uranium recovers its lost activity. 
 
 The rate at which UrX loses its activity was determined 
 by Rutherford and Soddy. The decay curve, like the decay 
 curves for simple radioactive products, is exponential, and 
 UrX loses half its activity in about 22 days. The recovery 
 curve of uranium measured by the (S rays, due to the fresh 
 production of UrX in the uranium, is complementary to the 
 decay curve. 
 
 From analogy with the corresponding results observed in 
 thorium and radium, we may thus conclude that uranium pro- 
 duces the new product UrX at a constant rate. Since the a 
 ray activity is unaffected by the removal of UrX, it seems 
 probable that the uranium atom breaks up with the emission 
 of an a particle and then becomes the atom of UrX. This 
 in turn breaks up with the expulsion of a IB particle. The 
 product resulting from the transformation of UrX is either 
 inactive, or active to such a feeble degree that its transformation 
 cannot be directly followed by the electric method. 
 
 The changes taking place in uranium are diagram matically 
 illustrated below. 
 
 1 Becquerel: Comptes rendus, cxxxi, p. 137 (1900) ; cxxxiii, p. 977 (1901). 
 
URANIUM AND ACTINIUM 163 
 
 a particle 
 
 Uranium atom y ray 
 
 \ 
 
 Atom of UrX > (3 particle 
 
 \ 
 
 It has been pointed out in the last chapter that UrX probably 
 undergoes one or more further changes of a long period, possi- 
 bly rayless in character, and is finally transformed into radium. 
 
 There are several points of interest in connection with the 
 ft ray activity exhibited by uranium. Meyer and Schweidler l 
 drew attention to some remarkable variations of the ft activity 
 of uranium during crystallization under various conditions. 
 This activity varies in a most capricious manner, as if the 
 process of crystallization had some direct effect on the rate 
 of transformation of UrX. Some later experiments made by 
 Dr. Godlewski 2 in the laboratory of the writer finally led to 
 a simple explanation of these puzzling phenomena observed by 
 Meyer and Schweidler. 
 
 Some uranium nitrate was heated and sufficient water added 
 for complete solution. A small dish containing the heated 
 solution was then placed under a ft ray electroscope. The ft 
 ray activity of the solution remained sensibly constant during 
 the cooling of the solution, but the moment crystallization com- 
 menced at the bottom of the dish, the ft ray activity increased 
 rapidly, and reached several times its initial value at the com- 
 pletion of the crystallization. After reaching a maximum, the 
 activity gradually diminished again, and about a week later 
 had reached a value equal to that of the uranium nitrate before 
 solution. 
 
 Another simple experiment was then made. A cake of 
 crystals so formed was removed from the dish immediately 
 after crystallization was completed, and inverted under the elec- 
 troscope. The ft ray activity was much less than for the 
 other side of the cake, and gradually increased again to the 
 
 1 Meyer and Schweidler: Wien Ber., cxiii, July, 1904. 
 
 2 Godlewski: Phil. Mag., July, 1905. ' 
 
164 RADIOACTIVE TRANSFORMATIONS 
 
 normal value. The explanation of this result is as follows. 
 UrX is more soluble in water than uranium itself. When the 
 crystallization starts at the bottom of the dish, the UrX is 
 pushed towards the surface of the solution. The $ rays enter- 
 ing the electroscope have on an average to pass through a less 
 depth of uranium than before. The ft ray activity will thus 
 increase until the crystallization is complete. The lower sur- 
 face of the plate of crystals will contain less than the normal 
 amount of UrX, and consequently will show a smaller ft ray 
 effect. The gradual decrease of the /3 ray activity of the 
 upper surface, and the increase of activity of the lower, appears 
 to be due to a diffusion of the UrX through the mass of 
 crystals. This process continues until the UrX is again uni- 
 formly distributed throughout the crystalline mass. This dif- 
 fusion takes place comparatively rapidly even in a plate of 
 completely dry crystals. An effect of this kind, which is quite 
 likely to occur in any mixture of products differing in solu- 
 bility, shows how much care is necessary in interpreting varia- 
 tions of activity in a mass of substance which has just been 
 subjected to chemical treatment. 
 
 The fact that UrX is more soluble in water than uranium 
 can be simply utilized to effect a partial separation of UrX. 
 If uranium nitrate is dissolved in a slight excess of water, the 
 liquid left on the surface after crystallization contains a large 
 fraction of the total amount of the UrX originally present in 
 the uranium. 
 
 CHANGES IN ACTINIUM 
 
 Shortly after the discovery of radium and polonium, Debierne 
 noted the presence of a new radioactive substance in pitch- 
 blende residues which he called actinium. This was removed 
 from the radioactive mineral with the thorium, but can be 
 separated from it by suitable methods. Very little was known 
 for several years about the radioactive peculiarities of this sub- 
 stance. In the meantime, Giesel had independently observed 
 that a new radioactive substance was removed with the lantha- 
 num and cerium present in the radioactive mineral. This 
 
URANIUM AND ACTINIUM 165 
 
 substance emitted very freely a short-lived emanation, and it 
 was for this reason that he first termed it the "emanating 
 substance " a name which was later changed to "emanium." 
 Debierne found that actinium gave out an emanation which lost 
 half its activity in 3.9 seconds. Later work by various ob- 
 servers has shown that the emanation and the excited activity 
 produced by emanium and actinium have the same rates of 
 decay. 
 
 The active constituent present in the actinium of Debierne 
 is thus identical with that in the "emanium " of Giesel, and the 
 original name "actinium " will in consequence be used for this 
 substance. Actinium has not yet been separated in a suffi- 
 ciently pure state to examine its atomic weight or spectrum. 
 
 Very active preparations of actinium have already been ob- 
 tained by Giesel and Debierne, and it seems probable that in 
 the pure state actinium will prove to be of the same order of 
 activity as radium. The emanation is given out very freely 
 from the preparations of Giesel, and excites phosphorescence 
 on a zinc sulphide screen brought near it. The phenomenon 
 of scintillations is shown by actinium rays to an even more 
 marked degree than by the a rays of radium. The continuous 
 and rapid emission of a short-lived emanation from actinium 
 can be simply illustrated by a very striking experiment. A 
 small quantity of actinium, enclosed in a paper envelope, is 
 placed on a zinc sulphide screen. The a particles emitted from 
 the mass of the substance are stopped by the paper, but the 
 emanation readily diffuses through it into the surrounding air. 
 The a particles expelled from the emanation produce luminosity 
 in the zinc sulphide screen. On examination with a lens, this 
 light is seen to be made up of a multitude of brilliant scintilla- 
 tions. A puff of air removes the emanation, and the luminosity 
 disappears for a moment, but returns almost immediately as a 
 fresh amount of emanation is supplied. The luminosity rapidly 
 spreads from the actinium over the screen by the process of 
 diffusion. The slightest current of air produces a marked wav- 
 ering effect on the luminosity and displaces the luminosity in 
 the direction of the air current. 
 
166 RADIOACTIVE TRANSFORMATIONS 
 
 Actinium gives out a, /3, and 7 rays. These radiations have 
 been examined by Godlewski. 1 The /3 rays are apparently 
 fairly homogeneous, and have less power of penetration than the 
 corresponding rays from other active substances. This shows 
 that the /3 particles are all projected at about the same velocity, 
 and that this velocity is less than that of the average /3 rays 
 from other substances. 
 
 The 7 rays also have much less penetrating power than those 
 from radium. It seems not unlikely that the absence of very 
 penetrating 7 rays is connected with the absence of swiftly 
 moving {3 particles, for it is probable that the {3 particle, which 
 is projected from radium with a velocity nearly that of light, 
 will give rise to a more penetrating pulse than one projected 
 at a much lower speed. 
 
 In radioactive properties, actinium shows a remarkable simi- 
 larity to thorium. It emits a short-lived emanation, and this 
 is transformed into an active deposit which is concentrated on 
 the negative electrode in an electric field. 
 
 The activity of the deposit obtained by a long exposure to 
 the emanation subsequently diminishes, and ten minutes after 
 removal from the emanation, decays exponentially with a period 
 of about 34 minutes. Miss Brooks 2 showed that the curves of 
 excited activity for a short exposure exhibited the same general 
 behavior as the corresponding curves obtained for the active 
 deposit of thorium. The activity at first increased, passed 
 through a maximum after about 8 minutes, and finally decayed 
 exponentially, with a period of 34 minutes. 
 
 These results admit of the same explanation as in the case of 
 the active deposit of thorium. The emanation which gives out 
 a rays changes into a rayless product, actinium A, which is 
 half transformed in 34 minutes. This changes into another 
 substance called actinium B, which is half transformed in 
 about 2 minutes, and emits a, /?, and 7 rays. 
 
 The choice of the 2 minute period for actinium B rather than 
 .for actinium A followed from an observation of Miss Brooks. 
 
 1 Godlewski: Phil. Mag., Sept., 1905. 
 
 2 Miss Brooks: Phil. Mag., Sept., 1904. 
 
URANIUM AND ACTINIUM 16T 
 
 The active deposit, obtained on a platinum plate, was dissolved 
 in hydrochloric acid. The solution was then electrolyzed, and 
 an active substance which emitted a rays was obtained on one 
 of the electrodes. This lost its activity exponentially with 
 the period of about 1.5 minutes. This result shows that 
 actinium B, which emits rays, must have the shorter period. 
 
 The analogy with thorium became still closer when God- 
 lewski 1 and Giesel 2 independently separated from actinium a 
 very active substance called actinium X. This was effected 
 by precipitation with ammonia in exactly the same way as is 
 required for the separation of ThX from thorium. The actin- 
 ium X after precipitation of the actinium remains behind in 
 the filtrate mixed with actinium A and B. Godlewski found 
 that actinium X lost its activity exponentially with a period of 
 about 10 days. The actinium, freed from actinium X, at the 
 same time recovered its activity. There are, however, several 
 interesting points of difference in the chemical separation of 
 actinium X and ThX from their respective elements. In the 
 case of thorium, thorium A and B are only slightly soluble in 
 ammonia," and consequently are not removed with the ThX. 
 Quite the reverse holds for actinium. The active deposit is 
 readily soluble in ammonia, and consequently is separated with 
 actinium X. 
 
 After removal of actinium X by successive precipitations, the 
 actinium itself retains only a small proportion of its normal 
 activity, while in the case of thorium, the residual a ray activity 
 is about one quarter of the total. It seems probable that, if the 
 actinium were completely freed from actinium X and its sub- 
 sequent products, the element itself would show no activity 
 measured by, the a or /3 rays, or, in other words, that actinium 
 itself is a rayless product. From the results of Hahn, dis- 
 cussed on page 168, it has already been pointed out that thorium 
 freed from radiothorium may also prove to be a rayless substance. 3 
 
 1 Godlewski: Phil. Mag., July, 1905. 
 
 2 Giesel: Jahrbuch. d. Radioaktivitat, i, p. 358 (1904). 
 
 3 Hahn (Nature, April 12, 1906) has recently separated another product from 
 actinium which he has called " radioactinium." This product is intermediate between 
 actinium and actinium X, emits a rays, and has a period of transformation of about 
 
168 RADIOACTIVE TRANSFORMATIONS 
 
 Godlewski showed that the emanation from actinium was a 
 direct product of actinium X, and not of actinium itself. In 
 this respect, ThX and actinium X have very similar properties. 
 The transformations taking place in actinium are shown in 
 Fig. 40. 
 
 On comparison of the changes taking place in actinium and 
 thorium (see Fig. 41) the similarity in the succession of changes 
 in the two substances is very noteworthy. Not only are the 
 products equal in number, but the corresponding products are 
 closely allied in general chemical and physical properties. The 
 active deposit of actinium differs somewhat from that of thorium 
 
 ACT//V/VAJ. ACTX- EMANATION. ACT A. 
 
 FIG. 40. 
 Actinium and its family of products. 
 
 in the ease with which it is dissolved by various solutions and 
 the lower temperature at which it is volatilized. 
 
 This similarity in the radioactive changes of the two sub- 
 stances indicates that the atoms of actinium and thorium, while 
 chemically distinct, are very similarly constituted, and that 
 when once the process of disintegration is started, the atom of 
 both substances passes through a similar succession of changes. 
 
 CONNECTION BETWEEN THE RADIOELEMENTS 
 
 The series of transformations taking place in the radioele- 
 ments are shown in Fig. 41. l 
 
 20 days. Actinium itself is a rayless product. Godlewski had unknowingly separated 
 this product from his actinium, for otherwise the actinium would have emitted a rays, 
 due to the presence of radioactinium. Levin has found that actinium X does not emit 
 rays. The rays from actinium arise only from the product actinium B. 
 
 1 In the diagram (Fig. 41), the products radiothorium, radioactinium, should be 
 introduced between thorium and thorium X, actinium and actinium X, respectively. 
 
URANIUM AND ACTINIUM 169 
 
 The substances thorium, radium, and actinium exhibit many 
 interesting points of similarity in the course of their transfor- 
 mation. Each gives rise to an emanation whose life is short 
 compared with that of the primary element itself. Such experi- 
 ments as have yet been made, indicate that these emanations 
 have no definite combining properties, but belong apparently 
 
 UF\ANIUM UK- X 
 
 i i s ".\ 
 
 6-6-6-0-& 
 
 THORIUM TH. X e/*A/V. TH.A TH.B 
 
 ACTTVC "*" O P06JT 
 
 ACTINIUM ACT. X EMAN- ACT. A ACT. S 
 
 ACT/VC DEPOSIT 
 
 ACT/ve DEPOSIT f^AflO CHAN,E ACT/Vfi DEPOSIT SLOW 
 
 FIG. 41. 
 The radioelements and their family of products. 
 
 to the helium-argon group of inert gases. In each case, the 
 emanation gives rise to a non-volatile substance which is depos- 
 ited on the surface of bodies and is concentrated on the nega- 
 tive electrode in an electric field. The changes in these active 
 deposits are also very similar, for each gives rise to a ray less 
 product, followed by a product which emits all three types of 
 rays. In each case, also, the rayless product has a longer period, 
 
170 RADIOACTIVE TRANSFORMATIONS 
 
 or, in other words, is a more stable substance than the ray 
 product which results from its transformation. 
 
 The disintegration of the corresponding products thorium B, 
 actinium B, and radium C is of a more violent character than 
 is observed in the other products, for not only is an a particle 
 expelled at a greater speed, but a /3 particle is also thrown 
 off at great velocity. After this violent explosion within the 
 atom, the resulting atomic system sinks into a more permanent 
 state of equilibrium, for the succeeding products thorium C and 
 actinium C have not so far been detected by radioactive methods, 
 while radium D is transformed at a very slow rate. 
 
 This similarity in the properties of the various families of 
 products is too marked to be considered a mere coincidence, 
 and indicates that there is some underlying law which governs 
 the successive stages of the disintegration of all the radioele- 
 ments. The transformation products mark the distinct stages 
 in the career of disintegration of the atoms, and represent the 
 halting places where the atoms are able to exist for an appre- 
 ciable time before again breaking up into other more or less 
 stable configurations. 
 
 The interesting question arises whether the atom after losing 
 an a particle is able to exist for a short time in more than one 
 stable form. After the expulsion of an a particle with explo- 
 sive violence, there must result a rearrangement of the parts of 
 the atom to form a permanently or temporarily stable system. 
 It is conceivable that more than one fairly stable arrangement 
 may be possible, and, in such a case, two or more products of 
 disintegration must be produced in addition to the expelled a 
 particles. These stable atomic systems, although of equal 
 atomic weights, would exhibit differences in chemical properties, 
 and it should be possible to separate them from one another. It 
 is not necessary that these products should be formed in equal 
 amount. One might exist in comparatively large amount 
 compared with the others. 
 
 There is in addition another possibility to be borne in mind. 
 The violent disturbance in the atom resulting in the expulsion 
 of an a particle may cause an actual breaking up of the main 
 
URANIUM AND ACTINIUM 171 
 
 atom into two parts, and thus give rise to an equal number of 
 atoms of different atomic weights in addition to the a particle. 
 For example, such an effect might arise during the violent 
 disintegration of radium C or thorium B. 
 
 So far, it has not been found necessary to choose between 
 these theories to explain the transformation products of the 
 different elements. The disintegration in each case results in 
 the appearance of only one substance in addition to the expelled 
 particles. It is not unlikely, however, that a still closer exami- 
 nation of the radioelements may show the existence of prod- 
 ucts which lie outside the main line of descent. The method 
 of electrolysis has already proved of great value in separating 
 products of the transformation of radioelements which are pres- 
 ent in infinitesimal amount in a solution, and its possibilities 
 in this direction are by no means exhausted. 
 
 RAYLESS TRANSFORMATIONS 
 
 We have seen that the great majority of the products break 
 up with the expulsion of an a particle; in addition, a small 
 number emit a ft particle with its accompaniment the 7 rays, 
 while a few emit only a ft particle. There is also a special 
 class of product which does not emit rays at all. 
 
 It has been shown that two of these rayless products exist 
 in radium and actinium, and probably two in thorium. The 
 method of showing the existence of such rayless products and of 
 determining their physical and chemical properties has already 
 been discussed in previous chapters. Since a rayless product 
 does not emit any ionizing type of radiation, its presence can 
 only be observed indirectly by examination of the variation in 
 the amount of the succeeding product. By such methods, we 
 are enabled not only to determine the period of change of the 
 rayless product, but also its more marked chemical and physical 
 properties. 
 
 These products are apparently similar in all respects to the 
 ray products, with the exception that there is no evidence of the 
 emission of a or ft particles. They are unstable substances 
 which break up according to the same law as the other active 
 
172 RADIOACTIVE TRANSFORMATIONS 
 
 products, and give rise to another substance of different phys- 
 ical and chemical properties. 
 
 There are two general ways of regarding the transformation 
 of a rayless product. In the first place, it may be supposed 
 that the transformation consists, not in an actual expulsion of 
 a part of the atomic system, but in a rearrangement of the 
 component parts of the atom to form a new temporarily stable 
 system. On such a view, the atom of the rayless product has 
 the same atomic weight as the succeeding product, but differs 
 from it so materially in atomic configuration that the physical 
 and chemical properties are quite distinct. The two products 
 may thus be considered to be somewhat analogous to the case 
 of an element like sulphur, which exists in two distinct forms. 
 This analogy is, however, only superficial, for the atoms of the 
 products possess entirely different chemical and physical proper- 
 ties whether in the solid state or in solution. 
 
 On the other hypothesis, the transformation of a rayless prod- 
 uct is supposed to be similar in character to that of a ray 
 product, the only difference being that the a particle is not 
 expelled with sufficient velocity to produce appreciable ioniza- 
 tion of the gas. There is an actual loss of mass during the 
 transformation, but this loss cannot be detected by the electric 
 method. In the light of some experimental results, discussed 
 in Chapter X, such an explanation appears not improbable. It 
 is there shown that when the velocity of the a particle falls 
 below about 40 per cent of the maximum velocity of the swift- 
 est a particle from radium, viz., that expelled from radium C, 
 the photographic, phosphorescent, and ionizing properties of the 
 a particle become relatively very small. Since the a particle 
 from radium C is projected with a velocity of about 1/15 the 
 velocity of light, it is seen that an a particle may be projected 
 from matter at a great speed, and yet produce a comparatively 
 weak electrical effect compared with that produced by a particle 
 projected at twice its velocity. The a particle from radium C 
 produces about 100,000 ions in the gas before it is absorbed, 
 and consequently the electrical effect due to the charged a par- 
 ticles alone would be insignificant in comparison with that due 
 
URANIUM AND ACTINIUM 173 
 
 to the ionization of the gas by the passage of the swift a particle 
 through it. Remembering that a rayless product is generally 
 followed by a product which emits high velocity a particles, 
 the strong ionization effect due to the latter would tend to 
 mask completely the small electrical effect due to the rayless 
 product alone, even if it emitted charged a particles at slow 
 velocity. 
 
 It is difficult to devise experiments to decide which of these 
 two hypotheses as to the nature of a rayless transformation is 
 correct, but the view that there is an expulsion of an a particle 
 at too low a velocity to be detected by ordinary means has many 
 points in its favor. 
 
 PROPERTIES OF THE PRODUCTS 
 
 We have seen that, with a few exceptions, the products of 
 transformation of the radioelements exist in too small quanti- 
 ties to be ever detected by direct measurement of their weight 
 or volume. Even though the products exist in infinitesimal 
 amount in the parent matter, the property of emitting ionizing 
 radiations serves not only to measure their rate of transforma- 
 tion, but also to deduce some of their physical and chemical 
 properties. 
 
 The electric method has been utilized as an accurate means 
 of qualitative and quantitative analysis of radioactive matter 
 which is present in extraordinarily small amount. The pres- 
 ence of 10~ n gram of a slowly changing substance like radium 
 can easily be observed, while in the case of more rapidly chang- 
 ing matter like the thorium emanation, one hundred millionth 
 of this small amount is readily detectable. In fact, as has been 
 previously pointed out, the electric method is easily capable 
 of showing the presence of radioactive matter in which only 
 one atom breaks up per second, provided that a high velocity 
 particle is expelled during the transformation. 
 
 With the aid of the electroscope, the range of possible appli- 
 cation of chemical methods of separation has been enormously 
 extended. It has been found that the ordinary methods of 
 chemical separation of substances, whether depending on dif- 
 
174 RADIOACTIVE TRANSFORMATIONS 
 
 ferences in solubility or volatility, or upon electrolysis, still 
 apply to matter existing in infinitesimal proportion. For the 
 detection of minute amounts of active matter, the electroscope 
 far transcends in delicacy the balance or even the spectroscope. 
 
 The study of radioactivity has thus indirectly furnished 
 chemistry with new methods of attack on the properties of 
 active matter existing in extremely small quantity. Much still 
 remains to be done in this new field of work whose importance 
 is not as yet sufficiently recognized. 
 
 Attention has already been drawn to the radical alteration in 
 properties of successive transformation products. This is well 
 exemplified by the transformation of radium into its emanation 
 and of the emanation into the active deposit. Each of these 
 substances is entirely dissimilar in physical and chemical nature 
 to the others, and, but for other evidence, it would be difficult 
 to believe that these substances were derived from the direct 
 transformation of the radium atom. 
 
 The atom at most stages of its disintegration loses an a 
 particle, which has an apparent mass about twice that of the 
 hydrogen atom. This decrease in the mass of the atom of 
 about one per cent gives rise to an entirely new atomic con- 
 figuration whose chemical and physical properties bear, as we 
 have seen, no obvious relation to the parent atom. This radical 
 change in the properties of the atom is not, however, very sur- 
 prising if we consider chemical analogies. Elements which do 
 not differ much in atomic weight often possess entirely dis- 
 similar properties, and thus we might reasonably expect that a 
 decrease of the atomic mass would result in a marked change 
 of the chemical and physical nature of the substance. 
 
 There cannot now be any doubt that the radioactive products 
 arise from the successive transformations of the atoms of matter, 
 and not of the molecules. Each transformation product is a 
 distinct element, which differs only from the well known in- 
 active elements in the comparative instability of the atoms com- 
 posing it. There can be no doubt, for example, that the radium 
 emanation, while it lasts, is a new elementary substance with an 
 atomic weight and spectrum that distinguish it from all other 
 
URANIUM AND ACTINIUM 175 
 
 elements. If, for instance, it were possible to examine chemi- 
 cally any one of the simple products in a time which is short 
 compared with its period of transformation, the substance would 
 be found to have all the distinctive properties of a new element. 
 It would possess a definite atomic weight and spectrum and other 
 distinctive physical and chemical properties. In regard to their 
 position as elements, no line of demarcation can be drawn 
 between the comparatively stable elements like uranium, tho- 
 rium, and radium, and their rapidly changing products. From 
 a radioactive point of view, the atoms of these substances differ 
 from one another mainly in stability. The atoms of each radio- 
 element may differ enormously in stability, but ultimately, if 
 sufficient time is allowed, all of these substances must be trans- 
 formed through a succession of stages and disappear. There 
 will finally remain only the inactive or stable products of their 
 decomposition. 
 
 There is no evidence that the process of disintegration, when 
 once started, is reversible under ordinary conditions. We can 
 obtain the radium emanation from radium, but cannot change 
 the emanation back again into radium. The question whether 
 this process has been reversible under some possible conditions 
 existing during the earth's history will be considered later in 
 Chapter IX. 
 
 LIFE OF THE RADIOELEMENTS 
 
 We have seen that every simple product which emits a 
 radiation decreases in amount on account of its transformation 
 into another substance. The rate of transformation is directly 
 porportional to the constant A, and inversely proportional to the 
 period of transformation. The period of transformation of any 
 simple product may be taken as a comparative measure of the 
 stability of the atoms composing it. It is at once seen that the 
 atomic stability of the products whose rate of transformation has 
 been directly measured varies over an enormous range. For ex- 
 ample, the atoms of radium F, which are half transformed in 
 140 days, are over three million times as stable as the atoms of 
 the actinium emanation which is half transformed in 3. 9 seconds. 
 
176 RADIOACTIVE TRANSFORMATIONS 
 
 This range of stability of the atoms is still further extended 
 when we include the atoms of the primary elements uranium 
 and thorium. 
 
 The periods of transformation of these substances can be 
 approximately deduced by comparison of their a ray activities. 
 Since uranium is the parent of radium, the relative amounts of 
 uranium and radium present in an old radioactive mineral are 
 directly proportional to the periods of transformation of the two 
 substances. Now it has been shown that 3.8 x 10~ 7 gram of 
 radium is present per gram of uranium in any radioactive 
 mineral. Since radium is half transformed in about 1800 
 
 10 7 
 years, uranium must be half transformed in 1300 x o~o, or about 
 
 3.4 x 10 9 years. 
 
 The period of transformation of thorium is probably three 
 or four times greater than this, since its activity is about the 
 same as that of uranium, but gives rise to four a ray products 
 to the one of uranium. In order that a large fraction of any 
 given mass of uranium may be transformed, a period of at least 
 ten thousand million years is necessary. 
 
 The period of transformation of actinium cannot be deter- 
 mined until it has been obtained in a pure state. If, however, 
 its activity is of the same order as that of radium, its period 
 will also be of the same order. 
 
 There appears to be no obvious relation existing between the 
 periods of the successive products nor between the periods of 
 the different families of products. It is a matter of remark, 
 however, that a substance of great stability is generally fol- 
 lowed by a number of comparatively unstable products. This is 
 well exemplified in the case of thorium, radium, and actinium, 
 where most of their known products suffer rapid transformation. 
 
 CONNECTION BETWEEN URANIUM, RADIUM, AND ACTINIUM 
 
 The connection that exists between uranium and radium, and 
 its products, radiolead and polonium, has been clearly brought 
 out, and it is of interest to examine whether any similar rela- 
 tion exists between uranium and thorium, and uranium and 
 
URANIUM AND ACTINIUM 177 
 
 actinium. The latter substance is always found in uranium 
 minerals, and since it probably has a radioactive life comparable 
 with that of radium, it must be produced in some way from the 
 parent substance uranium. 
 
 This question was examined by Dr. Boltwood and the writer. 
 If actinium, for example, was a product of uranium in the main 
 line of descent, the activity due to actinium or to a uranium 
 mineral should be comparable with that of radium. Since there 
 would be a state of equilibrium between the uranium and 
 actinium, the same number of atoms of each should break up 
 per second. Since actinium has four a ray products and 
 radium five, the activity due to actinium in the mineral should 
 be comparable with that due to radium and its family of 
 products. Experiment, however, showed that the activity 
 observed in Colorado uraninite, for example, was almost entirely 
 due to the uranium and radium contained in it. Its activity 
 due to actinium was certainly only a small portion of that due 
 to radium and its products. It seems probable that actinium 
 does arise from uranium, but that it is not a lineal descendant 
 of uranium in the same sense that radium is. It has already 
 been pointed out that in some of the transformations, two 
 distinct transition substances may be produced. It appears 
 likely that actinium will prove to be derived from uranium or 
 one of its products, but that it is produced in much less amount 
 than the other product. Such a relation would explain the con- 
 nection that apparently exists between uranium and actinium, 
 and at the same time would account for the small amount of 
 actinium present. 
 
 In regard to the connection between thorium and uranium, 
 the evidence is not very definite. Many minerals contain 
 uranium and very little thorium, but Strutt has shown that 
 every thorium mineral examined contains some uranium and 
 radium. Strutt has suggested that thorium is the parent of 
 uranium. Such a relation is suggested by an analysis of the 
 mineral thorianite. This mineral, of very great geological age, 
 is found in Ceylon, and contains about 70 per cent of thorium 
 and 12 per cent of uranium. The uranium in this mineral may 
 
 12 
 
178 RADIOACTIVE TRANSFORMATIONS 
 
 have been derived from the decomposition of the thorium. 
 There is, however, a serious objection to this view, for the 
 atomic weight of thorium, 232.5, is less than that of the usually 
 accepted atomic weight of uranium, 238.5. If these atomic 
 weights are correct, it does not appear likely that thorium is the 
 parent of uranium, unless the process of production of uranium 
 from thorium is very different from that usually observed in 
 radioactive transformations. 
 
CHAPTER VIII 
 
 THE PRODUCTION OF HELIUM FROM RADIUM AND THE 
 TRANSFORMATION OF MATTER 
 
 THE history of the discovery of helium possesses some fea- 
 tures of unusual dramatic interest. In 1868, Janssen and 
 Lockyer observed in the spectrum of the sun's chromosphere 
 a bright yellow line, which could not be identified with that 
 of any known terrestrial substance. Lockyer gave the name 
 " helium " to this supposed new element. Further comparison 
 showed that certain other spectral lines in the chromosphere 
 always accompanied the yellow line and were probably charac- 
 teristic of helium. 
 
 The spectrum of helium is observed not only in the sun, 
 but also in many of the stars ; and in some classes of stars, now 
 known as helium stars, the spectrum of helium predominates. 
 No evidence of the existence of helium on the earth was dis- 
 covered until 1895. Shortly after the discovery of argon in 
 the atmosphere, by Lord Rayleigh and Sir William Ramsay, a 
 search was made to see if argon could be obtained from mineral 
 sources. In 1895, Miers in a letter to " Nature " drew attention 
 to some results obtained by Hillebrande of the U. S. Geological 
 Survey in 1891. In the course of the detailed analysis of many 
 of the minerals containing uranium, a considerable quantity of 
 gas was found by Hillebrande 1 to be given off on solution of the 
 minerals. At the time he thought this gas was nitrogen, al- 
 though attention was drawn to some peculiarities of its behav- 
 ior as compared with ordinary nitrogen. The mineral clevite 
 especially gave off a large quantity of gas when heated or dis- 
 solved. Ramsay procured some of this mineral in order to see 
 whether this gas might prove to be argon. On introducing the 
 
 i Hillebrande, Bull. U. S. Geolog. Survey, No. 78, p. 43 (1891). 
 
180 RADIOACTIVE TRANSFORMATIONS 
 
 gas liberated from clevite into a vacuum tube, a spectrum was 
 observed entirely different from that of argon. 1 The spectrum 
 was carefully examined by Lockyer 2 and found to be identical 
 with that of the new element helium, previously discovered 
 by him in the sun. After a lapse of thirty years since its dis- 
 covery in the sun, helium had at last been found to exist in the 
 earth. An examination of the properties of helium soon fol- 
 lowed. It has a well-marked complex spectrum of bright lines, 
 of which the most noticeable is a bright yellow line D 3 close to 
 the sodium D lines. 
 
 It is a light gas about twice as dense as hydrogen, and, 
 excepting the latter, has a lighter atom than any other known 
 element. Like argon, it refuses to combine with any other sub- 
 stance, and must therefore be classed with the group of chemi- 
 cally inert gases discovered by Ramsay in the atmosphere. By 
 measurement of the velocity of sound in a tube filled with 
 helium, the ratio of its two specific heats was found to be 1.66. 
 The ratio for diatomic gases like hydrogen and oxygen is 1.41. 
 This result suggests that helium is monatomic, i. e., that the 
 helium molecule consists of only one atom ; or, in other words, 
 that the atom and molecule of helium are the same. Since the 
 density of helium was found to be 1.98 times that of hydrogen 
 at the same temperature and pressure, and since the hydrogen 
 molecule contains two atoms, it was concluded that the atomic 
 weight of helium is twice this amount, or 3.96. It must be 
 remembered that this atomic weight has been determined only 
 from density observations, since helium cannot be made to enter 
 into any chemical combination, and consequently the value 
 given for its atomic weight has not the same claim to accuracy 
 as the atomic weights of many of the other elements which have 
 been determined by more rigorous chemical methods. 
 
 Helium was found to exist in minute proportion in the atmos- 
 phere. In a recent paper Ramsay has concluded that one volume 
 of helium is present in 245,000 volumes of air. The occurrence 
 of helium in certain minerals was most remarkable, for there ap- 
 
 1 Ramsay, Proc. Roy. Soc., Iviii, p. 65 (1895). 
 
 2 Lockyer, Proc. Roy. Soc., Iviii, p. 67 (1895). 
 
THE PRODUCTION OF HELIUM 181 
 
 peared no obvious reason why an inert gaseous element should 
 be found associated with minerals, which in many cases are 
 impervious to the passage of water or gases. 
 
 Quite a new light was thrown on this subject as a result of 
 the discovery of radioactivity. On the disintegration theory of 
 radioactivity, it was to be expected that the final or inactive 
 products of the transformation of the radioelemeiits would be 
 found in the radioactive minerals. Since many of the radioac- 
 tive minerals are of extreme antiquity, it was reasonable to 
 suppose that the inactive products of the transformation of 
 radioactive substances, provided they did not escape, would be 
 found associated in some quantity with the radioactive matter 
 as its invariable companions. In looking for a possible disin- 
 tegration product, the occurrence of helium in all radioactive 
 minerals was noteworthy, for helium is mainly found in minerals 
 which contain a large quantity of uranium or of thorium. 
 
 For these and other reasons, Rutherford and Soddy 1 sug- 
 gested that helium might prove to be a disintegration product 
 of the radioelements. Additional weight was lent to this sug- 
 gestion by the writer's discovery that the a particle expelled from 
 radium has an apparent mass about twice that of the hydrogen 
 atom and might prove to be an atom of helium. 
 
 In the beginning of 1903, thanks to Dr. Giesel of Braun- 
 schweig, small quantities of pure radium bromide were placed 
 on the market. Ramsay and Soddy obtained 30 milligrams of 
 the bromide and proceeded to see if it were possible to detect 
 the presence of helium in the gases released from it. In the 
 first experiment the radium bromide was dissolved in water and 
 the accumulated gases drawn off. It was known that radium 
 bromide produced hydrogen and oxygen, and these gases were 
 removed by suitable methods. A small bubble of residual gas 
 was obtained which, on introduction into a vacuum tube, showed 
 the characteristic D 3 line of helium. 2 Using another and some- 
 what older sample of radium bromide, lent by the writer, the 
 
 1 Rutherford and Soddy: Phil. Mag., p. 582, 1902, pp. 453 and 579, 1903. 
 
 2 Ramsay and Soddy: Nature, July 16, p. 246, 1903. Proc. Roy. Soc., Ixxii, 
 p. 204 (1903); Ixxiii, p. 346 (1904). 
 
182 RADIOACTIVE TRANSFORMATIONS 
 
 residual gas released by solution of the radium was found to 
 give a complete spectrum of helium. 
 
 This experiment showed that helium was produced by radium 
 and retained to some extent in the solid compound. Further 
 experiments revealed a still more interesting fact. The emana- 
 tion from the 60 milligrams of radium bromide was condensed 
 in a glass tube and the other gases pumped out. After vola- 
 tilization, the emanation was introduced into a small vacuum 
 tube. The spectrum at first showed no sign of the helium lines, 
 but after three days the D 3 line of helium made its appearance, 
 and after five days the complete spectrum of helium was ob- 
 served. This experiment shows that helium is produced from 
 the emanation, for no evidence of its presence was obtained 
 immediately after the introduction of the emanation into the 
 spectrum tube. 
 
 The discovery of the production of helium by the radium 
 emanation was of great importance, as it showed in a striking 
 manner the extraordinary nature of the processes occurring in 
 radium, and was the first definite evidence of the possibility of 
 one element being transformed into another stable element. 
 The experiments were not easy of performance, as the helium 
 was present only in minute amount, and the experience gained 
 by Ramsay in his previous work on the rare gases in the atmos- 
 phere was of the greatest practical value in bringing the experi- 
 ments to such a successful conclusion. 
 
 The production of helium by radium has been confirmed by 
 a number of experimenters. Curie and Dewar 1 made in this 
 connection a most interesting experiment, which showed con- 
 clusively that the helium was produced from radium and could 
 not be ascribed to a possible occlusion of helium in the ra- 
 dium bromide. A large quantity of radium chloride was in- 
 troduced into a quartz tube and the radium heated to fusion. 
 The emanation and gases in the tube were pumped out and 
 the tube sealed. One month later, Deslandres examined the 
 spectrum of the gases in the tube by placing layers of foil 
 over the ends. A complete spectrum of helium was observed, 
 
 l Curie and Dewar: Comptes rendus, cxxxviii, p. 190 (1904). 
 
THE PRODUCTION OF HELIUM 183 
 
 showing that the helium had been produced from the radium 
 in the interval. 
 
 Recently Debierne l has found that helium is also produced 
 from active preparations of actinium. This result shows that 
 the helium must be a common product of these two substances, 
 which, from their radioactive and chemical behavior, must be 
 regarded as distinct elements. 
 
 THE POSITION OF HELIUM AS A TRANSFORMATION 
 PRODUCT OF RADIUM 
 
 We have already seen that radium is transformed through a 
 long succession of products, each of which has some distinctive 
 physical and chemical properties and a definite period of trans- 
 formation. These products differ from the ordinary chemical 
 elements only in the instability of their component atoms. 
 They must be regarded as transition elements with a limited life, 
 which break up into new forms of matter at a rate independent 
 of our control. 
 
 The distinction, however, between helium as a product of 
 radium and the family of transition products is mainly one of 
 atomic stability. As far as we know, helium is a stable element 
 which does not disappear, but in the case of all the radioactive 
 products, including the primary sources uranium and thorium, 
 the atoms are undoubtedly unstable. 
 
 It is now necessary to consider the position of helium as a 
 transformation product of radium. Some have considered that 
 helium is the end or final product of the disintegration of the 
 radium atom, but for this there is no experimental evidence. 
 We have seen that after the first rapid changes in the active 
 deposit of the emanation, there is produced a very slow transition 
 substance, radium D. If helium were the final product of the 
 transformation of the radium atom, the amount produced from 
 the emanation in the course of a few days would be infinitesi- 
 mally small. In addition there can be little doubt that the final 
 active product of radium, viz. radium F (polonium), is an ele- 
 ment of high atomic weight. 
 
 1 Debierne: Comptes rendus, cxli, p. 383 (1905). 
 
184 RADIOACTIVE TRANSFORMATIONS 
 
 The evidence, on the other hand, points strongly to the con- 
 clusion that the helium is formed by the a particles continuously 
 shot out from radium and its products. We shall see later 
 (Chapter X) that the experimental evidence shows that the a 
 particle shot out from the different a ray products of radium 
 has in each case the same mass, but varies in velocity for the 
 different products. 
 
 From observation of the deflection of the rays both in a strong 
 magnetic and a strong electric field, the velocity, and the value 
 e/m the ratio of the charge of the a particle to its mass 
 have been accurately measured. The ratio e/m is very nearly 
 5 x 10 3 . Now the ratio e/m for the hydrogen atom liberated in 
 the electrolysis of water is known to be 10 4 . If we assume that 
 the a particle carries the same charge as the hydrogen atom, the 
 mass of the a particle is twice that of the hydrogen atom. We 
 are here unfortunately confronted with several possibilities be- 
 tween which it is at present difficult to make a definite choice. 
 
 The value of e/m for the a particle may equally well be 
 explained on the assumptions that the a particle is (1) a mole- 
 cule of hydrogen, (2) a helium molecule carrying twice the 
 charge of the hydrogen atom, or (3) one half of the helium 
 molecule carrying the usual ionic charge. 
 
 The hypothesis that the a particle is a molecule of hydrogen 
 seems for many reasons improbable. If hydrogen is a constitu- 
 ent of the atoms of radioactive matter, it is to be expected that 
 it would be expelled in the atomic and not in the molecular 
 state. In all cases so far examined, when hydrogen is the car- 
 rier of an electric charge, the value of e/m is 10 4 . This is the 
 value to be expected for the hydrogen atom. For example, 
 Wien found that the maximum value of ejm for the canal rays 
 or positive ions, which are produced in an exhausted vacuum 
 tube, was 10 4 . In addition, it seems improbable that, even if 
 the hydrogen were projected initially in the molecular state, it 
 would escape decomposition into its component atoms in passing 
 through matter. 
 
 When it is remembered that an a particle is projected 
 with a velocity of about 12,000 miles per second, and collides 
 
THE PRODUCTION OF HELIUM 185 
 
 with every molecule in its path, the disturbance set up in the 
 molecule by the collisions must be very intense, and must tend 
 to rupture the bonds that hold the atoms of the molecule 
 together. Indeed, it seems very unlikely that the hydrogen mole- 
 cule under such conditions could escape decomposition into its 
 component atoms. If the a particle were a hydrogen molecule, 
 a considerable amount of free hydrogen should be present in 
 old radioactive minerals which are sufficiently dense to prevent 
 its escape. This does not appear to be the case, although in 
 some minerals there is a considerable quantity of water. On 
 the other hand, the comparatively large amount of helium pres- 
 ent supports the view that the a particle is connected with 
 helium. A strong argument in favor of the view of a connection, 
 between helium and the a particle rests on the observed facts that 
 helium is produced by actinium as well as by radium. The 
 only point of similarity between these two substances lies in the 
 expulsion of a particles. The production of helium by both 
 substances is at once obvious if the helium is derived from the 
 accumulated a particles, but is difficult to explain on any other 
 hypothesis. We are thus reduced to the view that either the a 
 particle is a helium atoni carrying twice the ionic charge, or 
 that it is half of a helium atom carrying an ionic charge. 
 
 The latter assumption involves the conception that helium, 
 while behaving as a chemical atom under ordinary chemical and 
 physical conditions, may exist in a still more elementary state as 
 a component of the atoms of radioactive matter, and that, after 
 expulsion, the a particles lose their charge and recombine to 
 form atoms of helium. 
 
 While such a view cannot be dismissed as inherently improb- 
 able, there is no direct evidence in its favor. On the other 
 hand, the second hypothesis has the merit of greater simplicity 
 and probability. 
 
 On this view, the a particle is in reality an atom of helium 
 which is either expelled with a double ionic charge or acquires 
 this charge in its passage through matter. Even if the a particle 
 were initially projected without a charge, it would almost certainly 
 acquire one after the first few collisions with the molecules in its 
 
186 RADIOACTIVE TRANSFORMATIONS 
 
 path. We know that the a particle is a very efficient ionizer, and 
 there is every reason to suppose that it would itself be ionized by 
 its collision with the molecules in its path, i. ., it would lose an 
 electron, and would consequently itself retain a positive charge. 
 
 If the a particle can remain stable with the loss of two elec- 
 trons, these electrons would almost certainly be ejected as a 
 result of the intense disturbance arising from the collision of 
 the a particle with the molecules in its path. The a particle 
 would then have twice the ordinary ionic charge, and the value 
 of e/m, as found by measurement, would be quite consistent 
 with the view that the a particle is an atom of helium. 
 
 If this be the case, the actual number of a particles expelled 
 from radium would be only one half of that deduced on the as- 
 sumption that the a particle carries a single charge. This would 
 make the rate of disintegration of radium only half of that cal- 
 culated in Chapter VI, and would consequently double its life. 
 
 In a similar way this assumption would reduce the calculated 
 volume of the emanation released from one gram of radium from 
 .8 c. mms. to .4 c. mms. This is smaller than the experimental 
 value about 1 c. mm. determined by Ramsay and Soddy, 
 but is of the right order of magnitude. 
 
 On the above assumptions, the volume of helium produced per 
 year per gram of radium can readily be calculated. If each a 
 particle carries twice the ionic charge, experiment shows that 
 1.25 X 10 11 a particles are expelled per second from one gram 
 of radium in equilibrium. The number expelled per year is 
 4.0 x 10 18 . Since one cubic centimetre of a gas at standard press- 
 ure and temperature contains 3.6 x 10 19 molecules, the volume 
 of helium produced per year per gram of radium is .11 c. cms. 
 
 Ramsay and Soddy made an estimate of the rate of produc- 
 tion of helium from radium in the following manner. The 
 helium produced from 50 mgrs. of radium bromide kept in a 
 closed vessel for 60 days was introduced into a vacuum tube. 
 Another similar tube was placed in series with it and the 
 amount of helium in the latter was adjusted until the discharge, 
 passed in series through the two tubes, showed the helium 
 lines with about the same intensity. In this way, they deduced 
 
THE PRODUCTION OF HELIUM 187 
 
 the volume obtained from the radium to be 0.1 cubic mm. 
 This corresponds to a rate of production of helium per gram of 
 radium per year of about 20 cubic mms. This is only about one 
 fifth of the theoretical amount calculated above. Ramsay and 
 Soddy do not lay much stress on the accuracy of their estimate, 
 as they consider that the presence of a trace of argon may have 
 seriously interfered with the correctness of the estimate by the 
 spectroscopic method. An accurate measurement of the rate of 
 production of helium by radium would be of the utmost value 
 at the present time in settling the connection between the a 
 particle and helium. 
 
 If the a particle is a helium atom, the greater proportion of 
 the a particles expelled from the emanation enclosed in a small 
 tube will be projected into the glass envelope. The swiftest 
 moving particles, viz., those expelled from radium C, would 
 probably penetrate the glass to a depth of about 1/20 of a 
 millimetre, while the slower moving particles would be stopped 
 after traversing a somewhat shorter distance. 
 
 It has already been pointed out (page 88) that this may 
 explain why the volume of the emanation in the first ex- 
 periment by Ramsay and Soddy shrank almost to zero. The 
 helium in this case was retained in the glass. In the second 
 experiment the helium may have diffused from the glass tube 
 into the gas again. Ramsay and Soddy endeavored to settle 
 this point by testing whether helium was released by heating a 
 glass tube in which the emanation had been enclosed for several 
 days and then removed. The spectroscope momentarily showed 
 some of the helium lines, but these were soon obscured by the 
 presence of other gases liberated by the heating of the tube. 
 
 AGE OF RADIOACTIVE MINERALS 
 
 The helium observed in the radioactive minerals is almost 
 certainly due to its production from the radium and other 
 radioactive substances contained therein. If the rate of pro- 
 duction of helium from known weights of the different radio- 
 elements were experimentally known, it should thus be possible 
 to determine the interval required for the production of the 
 
188 RADIOACTIVE TRANSFORMATIONS 
 
 amount of helium observed in radioactive minerals, or, in other 
 words, to determine the age of the mineral. This deduction is 
 based on the assumption that some of the denser and more com- 
 pact of the radioactive minerals are able to retain indefinitely a 
 large proportion of the helium imprisoned in their mass. In 
 many cases the minerals are not compact but porous, and under 
 such conditions most of the helium will escape from its mass. 
 Even supposing that some of the helium has been lost from 
 the denser minerals, we should be able to fix with some cer- 
 tainty a minimum limit for the age of the mineral. 
 
 In the absence of definite experimental data on the rates of pro- 
 duction of helium by the different radioelements, the deductions 
 are of necessity somewhat uncertain, but will nevertheless serve 
 to fix the probable order of the ages of the radioactive minerals. 
 
 It has already been pointed out that all the a particles ex- 
 pelled from radium have the same mass. In addition it has 
 been experimentally found that the a particle from thorium B 
 has the same mass as the a particle from radium. This would 
 suggest that the a particles projected from all radioactive sub- 
 stances have the same mass, and thus consist of the same kind 
 of matter. If the a particle is a helium atom, the amount of 
 helium produced per year by a known quantity of radioactive 
 matter can readily be deduced on these assumptions. 
 
 The number of products which expel a particles are now well 
 known for radium, thorium, and actinium. Including radium F, 
 radium has five a ray products, thorium five, and actinium four. 
 With regard to uranium itself, there is not the same certainty, 
 for only one product, UrX, which emits only /3 rays, has so far 
 been chemically isolated from uranium. The a particles appar- 
 ently are emitted by the element uranium itself ; at the same 
 time, there is some indirect evidence in support of the view that 
 uranium contains three a ray products. For the purpose of 
 calculation, we shall, however, assume that in uranium and 
 radium in equilibrium, one a particle is expelled from the 
 uranium for five from the radium. 
 
 Let us now consider an old uranium mineral which contains 
 one gram of uranium, and which has not allowed any of the 
 
THE PRODUCTION OF HELIUM 189 
 
 products of its decomposition to escape. The uranium and 
 radium are in radioactive equilibrium and 3.8 x 10~ 7 grams of 
 radium are present. For one a particle emitted by the ura- 
 nium, five are emitted by the radium and its products, including 
 radium F. Now we have shown that radium with its four a 
 ray products probably produces .11 c.c. of helium per gram per 
 year. The rate of production of helium by the uranium and ra- 
 dium in the mineral will consequently be J x .11 x 3.8 X 10 ~ 7 
 5.2 x 10~ 8 c.c. per year per gram of uranium. 
 
 Now, as an example of the method of calculation, let us con- 
 sider the mineral fergusonite which was found by Ramsay and 
 Travers to evolve 1.81 c.c. of helium per gram. The fergusonite 
 contains about 7 per cent of uranium. The amount of helium 
 contained in the mineral per gram of uranium is consequently 
 26 c.c. 
 
 Since the rate of production of helium per gram of uranium 
 and its radium products is 5.2 x 10~ 8 c.c. per year, the age of 
 the mineral must be at least 26 -f- .2 x 10~ 8 J years or 500 
 million years. This, as we have pointed out, is a minimum 
 estimate, for some of the helium has probably escaped. 
 
 We have assumed in this calculation that the amount of 
 uranium and radium present in the mineral remains sensibly 
 constant over this interval. This is approximately the case, 
 for the parent element uranium probably requires about 1000 
 million years to be half transformed. 
 
 As another example, let us take a uranium mineral obtained 
 from Glastonbury, Connecticut, which was analyzed by Hille- 
 brande. This mineral was very compact and of high density, 
 9.62. It contained 76 per cent of uranium and 2.41 per cent of 
 nitrogen. This nitrogen was almost certainly helium, and 
 dividing by seven to reduce to helium this gives the percentage 
 of helium as 0.344. This corresponds to 19 c.c. of helium per 
 gram of the mineral, or 25 c.c. per gram of uranium in the min- 
 eral. Using the same data as before, the age of the mineral 
 must be certainly not less than 500 million years. Some of the 
 uranium and thorium minerals do not contain much helium. 
 Some are porous, and must allow the helium to escape readily. 
 
190 RADIOACTIVE TRANSFORMATIONS 
 
 A considerable quantity of helium is, however, nearly always 
 found in the compact primary radioactive minerals, which from 
 geologic data are undoubtedly of great antiquity. 
 
 Hillebrande made a very extensive analysis of a number of 
 samples of minerals from Norway, North Carolina, and Connec- 
 ticut, which were mostly compact primary minerals, and noted 
 that a striking relation existed between the proportion of 
 uranium and of nitrogen (helium) that they contained. This 
 relation is referred to in the following words : 
 
 " Throughout the whole list of analyses in which nitrogen 
 (helium) has been estimated, the most striking feature is the 
 apparent relation between it and the UO 2 . This is especially 
 marked in the table of Norwegian uraninites, recalculated from 
 which the rule might almost be formulated that, given either 
 nitrogen or UO 2 , the other can be found by simple calculation. 
 The same ratio is not found in the Connecticut varieties, but 
 if the determination of nitrogen in the Branchville mineral is to 
 be depended on, the rule still holds that the higher the UO 2 the 
 higher likewise is the nitrogen. The Colorado and North Caro- 
 lina minerals are exceptions, but it should be borne in mind that 
 the former is amorphous, like the Bohemian, and possesses the 
 further similarity of containing no thoria, although zirconia 
 may take its place, and the North Carolina mineral is so much 
 altered that its original condition is unknown." 
 
 Very little helium, however, is found in the secondary radio- 
 active minerals, i. e., minerals which have been formed as a result 
 of the decomposition of the primary minerals. These minerals, 
 as Boltwood has pointed out, are undoubtedly in many cases of 
 far more recent formation than the primary minerals, and conse- 
 quently it is not to be expected that they should contain as 
 much helium. One of the most interesting deposits of a second- 
 ary uraninite is found at Joachimsthal in Bohemia, from which 
 most of our present supply of radium has been obtained. This 
 is rich in uranium, but contains very little helium. 
 
 When the data required for these calculations are known with 
 more definiteness, the presence of helium in radioactive minerals 
 will in special cases prove a most valuable method of computing 
 
THE PRODUCTION OF HELIUM 
 
 191 
 
 their probable age, and indirectly the probable age of the geo- 
 logical deposits in which the minerals are found. Indeed, it ap- 
 pears probable that it will prove one of the most reliable methods 
 of determining the age of the various geological formations. 
 
 SIGNIFICANCE OF THE PRESENCE OF LEAD IN 
 RADIOACTIVE MINERALS 
 
 If the a particle is a helium atom, the atomic weights of the suc- 
 cessive a ray products of radium must differ by equal steps of four 
 units. Now we have seen that uranium itself probably contains 
 three a ray products. Since the atomic weight of uranium is 
 238.5, the atomic weight of the residue of the uranium after the 
 expulsion of three a particles would be 238.5 12, = 226.5. 
 This is very close to the atomic weight of radium 225, which we 
 have seen is produced from uranium. Now radium emits five a 
 ray products altogether, and the atomic weight of the end prod- 
 uct of radium should be 238.5 32, = 206.5. This is very close 
 to the atomic weight of lead, 206.9. This calculation suggests 
 that lead may prove to be the final product of the decomposition 
 of radium, and this suggestion is strongly supported by the 
 observed fact that lead is always found associated with the 
 radioactive minerals, and especially in those primary minerals 
 which are rich in uranium. 
 
 The possible significance of the presence of lead in radioactive 
 minerals was first noted by Boltwood, 1 who has collected a large 
 amount of data bearing on this question. 
 
 The following table shows the collected results of an analysis 
 of different primary minerals made by Hillebrande : 
 
 Locality. 
 
 Percentage of 
 uranium. 
 
 Percentage of 
 lead. 
 
 Percentage of 
 nitrogen. 
 
 Glastonbury, Connecticut 
 Branchville, Connecticut 
 North Carolina .... 
 Norway . . . 
 
 70-72 
 
 74-75 
 77 
 56-66 
 
 3.07-3.26 
 4.35 
 4.20-4.53 
 7 62-13 87 
 
 2.41 
 
 2.63 
 
 1.03-1.28 
 
 Canada . 
 
 65 
 
 1049 
 
 086 
 
 
 
 
 
 Boltwood: Phil. Mag., April, 1905 ; Amer. Journ. Science, Oct., 1905. 
 
192 RADIOACTIVE TRANSFORMATIONS 
 
 Five samples were taken of the minerals from Glastonbury, 
 three from Branch ville, two from North Carolina, seven from 
 Norway, and one from Canada. In minerals obtained from the 
 same locality, there is a comparatively close agreement between 
 the amounts of lead contained in them. If helium and lead are 
 both products of the decomposition of the uranium radium miner- 
 als, there should exist a constant ratio between the percentage 
 of lead and helium in the minerals. The percentage of helium is 
 obtained from the above table by dividing the nitrogen percentage 
 by seven. Since probably .eight a particles are emitted from the 
 decomposition of uranium and radium for the production of one 
 
 8x4 
 atom of lead, the weight of helium formed should be ^7^-^ =.165 
 
 of the weight of lead. This is based on the assumption that all 
 the helium formed is imprisoned in the minerals. The ratio 
 actually found is about .11 for the Glastonbury minerals, .09 for 
 the Branch ville minerals, and about .016 for the Norway min- 
 erals. It will be noted that in all cases the ratio of helium to 
 lead is less than the theoretical ratio, indicating that in some 
 cases a large proportion of the helium formed in the mineral has 
 escaped. In the case of the Glastonbury minerals, the observed 
 ratio is in good agreement with theory. 
 
 If the production of lead from radium is well established, the 
 percentage of lead in radioactive minerals should be a far more 
 accurate method of deducing the age of the mineral than the 
 calculation based on the volume of helium, for the lead formed 
 in a compact mineral has no possibility of escape. 
 
 While the above considerations are of necessity somewhat 
 conjectural in the present state of our knowledge, they are of 
 value as indicating the possible methods of attacking the ques- 
 tion as to the final products of the decomposition of the radio- 
 active minerals. From a study of the data of analyses of 
 radioactive minerals, Boltwood has suggested that argon, hy- 
 drogen, bismuth, and some of the rare earths possibly owe 
 their origin to the transformation of the primary radioactive 
 substances. 
 
 It does not appear likely that we shall be able for many years 
 
THE PRODUCTION OF HELIUM 193 
 
 to prove or disprove experimentally that lead is the final prod- 
 uct of radium. In the first place, it is difficult for the experi- 
 menter to obtain sufficient radium for working material, and, in 
 the second place, the presence of the slowly transformed product 
 radium D makes a long interval necessary before lead will ap- 
 pear in appreciable quantity in the radium. A more suitable 
 substance with which to attack the question would be radium F 
 (radiotellurium) or radiolead (radium D). 
 
 CONSTITUTION OF THE RADIOELEMENTS 
 
 The view that the a particle is a helium atom suggests that 
 the atoms of uranium and radium are built up in part of atoms of 
 helium. If the final product of radium is lead, the radium atom 
 could thus be represented by the equation, Ra = Pb He^ while 
 Ur = Pb- ffe s . 
 
 It must be borne in mind, however, that these compounds of 
 helium are very different from ordinary molecular compounds. 
 Both radium and uranium behave as elementary substances, 
 which cannot be broken up by the application of physical or 
 chemical forces at our command. These substances spontane- 
 ously break up at a rate that is independent of known agencies, and 
 the disintegration is accompanied by the expulsion of a helium 
 atom with enormous velocity. The energy liberated in the form 
 of the kinetic energy of the expelled helium atoms is of quite a 
 different order from that observed in molecular reactions, being at 
 least one million times as great as that released in the most vio- 
 lent chemical combinations. It seems probable that the helium 
 atoms are in very rapid motion within the atoms of uranium 
 and radium, and for some reason escape from the atoms with the 
 velocities which they possessed in their orbits. The forces that 
 hold the helium atoms in place in the atom of the radioelements 
 are so strong that no means at our disposal are able to effect 
 their separation. 
 
 It seems probable that the a particles from thorium and actin- 
 ium are also helium atoms, so that these substances must also 
 be considered as compounds of some unknown substances with 
 
 13 
 
194 RADIOACTIVE TRANSFORMATIONS 
 
 helium. Five a ray products are known to exist in thorium, 
 and this would make the atomic weight of the residue of the 
 thorium atom 232.5 5x4, or 212.5. The nearest known 
 atomic weight to this is that of bismuth, viz. 208, and if 
 thorium should lose six a particles instead of five, the atomic 
 weight of the residue would be very nearly that of bismuth. This 
 substance also fulfils the condition required for a transformation 
 product of radioactive substances, for it is found in radioactive 
 minerals, although only in small amount compared with that 
 of lead in the old uranium minerals, -where little thorium is 
 present. 
 
 It thus appears that helium plays a most important part in the 
 constitution of the radioelements, and it is not unlikely that 
 helium as well as hydrogen may ultimately prove to be one of 
 the more elementary units of which the heavy atoms are built up. 
 In this connection it may prove more than a coincidence that a 
 number of the atomic weights of the elements differ by nearly 
 four units or by multiples of four units. 
 
 Many of the primary radioactive minerals were undoubtedly 
 deposited at the surface of the earth 100 to 1000 million years 
 ago, and since that time have been undergoing slow trans- 
 formation. There is no evidence at hand that this process of 
 degradation of matter is reversible under ordinary conditions 
 at the surface of the earth. It seems, however, reasonable to 
 suppose that under some conditions, existing possibly early in 
 the earth's history, the converse process took place, and that the 
 heavy atoms were built up from the lighter and more elementary 
 substances. 
 
 It may happen that the conditions for the formation of heavy 
 atoms may be found at the high pressures and temperatures 
 existing deep in the earth. It has been suggested to me by Dr. 
 Barrel, of Yale University, that the gradual building up of the 
 heavy and more complex atoms of matter may be slowly taking 
 place m the interior of the earth, and that this might possibly 
 account for the undoubtedly high density of the matter in the 
 interior of the earth, and also for the gradual shrinking of the 
 earth as a whole. 
 
THE PRODUCTION OF HELIUM 
 
 195 
 
 While such suggestions are at present highly speculative, it 
 appears not unreasonable to suppose that the formation of the 
 radioactive matter may still be in progress deep in the earth, 
 and that the radioactive deposits found at the surface to-day 
 have been forced up from below in past ages. 
 
 OF THE 
 
 ' 
 
 UNIVERSITY 
 
 OF 
 
 'FORN\^ 
 
CHAPTER IX 
 RADIOACTIVITY OF THE EARTH AND ATMOSPHERE 
 
 WE shall in this chapter briefly discuss the present state of 
 our knowledge of the radioactive condition of the earth and 
 atmosphere, and the possible bearing of the facts so far obtained 
 on the problems of the electrical state of the atmosphere and on 
 the internal heat of the earth. 
 
 ATMOSPHERIC RADIOACTIVITY 
 
 The remarkable development during the last few years of our 
 knowledge of the radioactive and electrical state of the atmos- 
 phere is one of unusual interest, and although the interval for 
 investigation has been short, a great deal of new and important 
 information has been accumulated. 
 
 Nearly a century ago, Coulomb and others drew attention to 
 the fact that a charged conductor placed inside a closed vessel 
 lost its charge more rapidly than could be explained by the con- 
 duction of electricity along the insulating support. This was 
 thought by Coulomb to be due to the molecules of air receiving 
 a charge from contact with the charged rod and then being 
 repelled from it. As early as 1850, Matte ucci observed that the 
 rate of loss of charge was independent of the potential of the 
 charged body. By using insulators of quartz rods of different 
 lengths and cross section, Boys in 1889 came to the conclusion 
 that the loss of charge could not be explained by the imperfect 
 insulation of the supports. 
 
 Shortly after science had become familiar with the ionization 
 of gases by X-rays and uranium rays, the question of the cause 
 of this loss of charge was independently attacked by Geitel 1 
 and C. T. R. Wilson, 2 using specially designed electroscopes to 
 
 1 Geitel: Physik. Zeit., ii, p. 116 (1900). 
 
 2 Wilson: Proc. Camb. Phil. Soc. : xi, p. 32 (1900); Proc. Roy. Soc., Ixviii, 
 p. 151 (1901). 
 
RADIOACTIVITY OF THE EARTH 197 
 
 measure the rate of discharge of a charged body inside a closed 
 vessel. Both came to the conclusion that the gradual loss of 
 charge was mainly due to an ionization of the air inside the 
 closed vessel. Above a certain voltage, the rate of discharge 
 was independent of the voltage, a result to be expected if the 
 ionization was very weak. It was at first thought that this 
 ionization in the gas was spontaneous and a property of the 
 gas itself, but later work has modified this conclusion. It is 
 now certain that a large part of the ionization observed in a 
 clean metal vessel results mainly from the emission of ionizing 
 radiations from its walls. A part is due to a very penetrating 
 radiation of the 7 ray type which is everywhere present on the 
 surface of the earth. The amount of ionization of a gas inside a 
 
 o 
 
 closed vessel depends on the nature and pressure of the gas and 
 of the material of the vessel. In most cases the ionization falls 
 off nearly proportionally with the pressure, and is approximately 
 proportional to the density of the gas. Both of these results 
 are to be expected if the ionization observed is due to radiations 
 from the walls or to a penetrating type of radiation passing 
 from the outside through the material of the vessel. 
 
 It must be borne in mind that the natural ionization observed 
 in closed vessels is extraordinarily minute, and special precau- 
 tions are usually necessary to measure it with accuracy. As- 
 suming that the ionization in a small silvered glass vessel was 
 uniform throughout its volume, C. T. R. Wilson found that not 
 more than 30 ions were produced per second per cubic centi- 
 metre of the enclosed air. In a vessel of one litre capacity, the 
 number of ions produced per second would be 30,000, or only 
 about one third of the total number of ions produced in air by 
 a single a particle emitted from radium. The expulsion of a 
 single a particle per second from the walls of the vessel would 
 thus more than account for the ionization observed. 
 
 After examining the discharge of electricity produced by air 
 in closed vessels, Elster and Geitel turned their attention to the 
 external air. They found that a charged body freely exposed to 
 the open air lost its charge far more rapidly than when placed in 
 a small closed vessel. Both positive and negative electricity is 
 
198 RADIOACTIVE TRANSFORMATIONS 
 
 discharged, but generally at unequal rates, a positively charged 
 body losing its charge somewhat more slowly than a negatively 
 charged one. The ionization of the open air was examined by 
 means of a portable electroscope. An insulated wire gauze was 
 connected to the charged electroscope, and the rate of loss of 
 charge of the electroscope was taken as a comparative measure 
 of the number of ions in the air. 
 
 In the course of their experiments on closed vessels, Elster 
 and Geitel noted that the rate of discharge increased for several 
 hours after the introduction of fresJa^jn?- Such a result was 
 known to occur when the radium or thorium emanation was 
 mixed with the air. This led them to try a bold experiment 
 to see if it were possible to extract a radioactive substance 
 from the atmosphere. The writer had shown that a negatively 
 charged wire exposed in the presence of the thorium emanation 
 became strongly active. This experiment suggested the method 
 of attacking the question. 1 A long wire was suspended on 
 insulating supports outside the laboratory and charged nega- 
 tively to a high potential by means of a static machine. After 
 some hours the wire was removed and coiled round the top of 
 an electroscope. There was an undoubted increase in its rate 
 of discharge, showing that the wire possessed the new property 
 of ionizing the gas. The effect died away after a time, and was 
 small after a few hours' interval. 
 
 Further experiments showed that the wire had been made 
 temporarily radioactive by exposure to the open air. The 
 amount of activity observed was independent of the nature of 
 the material of the wire, and in this respect the activity behaved 
 quite similarly to the excited activity imparted to bodies in the 
 presence of the radium and thorium emanations. 
 
 The active matter could be dissolved from the wire by 
 rubbing it with leather soaked in ammonia. In this way 
 an active substance was obtained capable of affecting a pho- 
 tographic plate through .1 mm. of aluminium and of produc- 
 ing weak phosphorescence on a screen of platinocyanide of 
 barium. 
 
 i Elster and Geitel: Physik. Zeit., iii, p. 76 (1901). 
 
RADIOACTIVITY OF THE EARTH 199 
 
 Rutherford and Allan l showed that similar activity could be ob- 
 tained from the open air in Montreal. The radiations consisted 
 of a and /3 rays, the former being responsible for most of the 
 ionization observed with bare wires. The activity of a wire made 
 active by exposure to the atmosphere decayed at about the same 
 rate as that of a wire made active by exposure to the radium 
 emanation. 
 
 Bumstead and Wheeler 2 examined the radioactive state of the 
 air at New Haven, and from a comparison of the rate of decay 
 of the active wire with that of a wire made active by exposure 
 to the radium emanation, showed conclusively that the activity 
 observed in the air in that locality was mainly due to the radium 
 emanation. A wire made active in the open air showed the 
 initial rapid drop of activity due to radium A, and the curve 
 of decay was identical with that due to the excited activity of 
 radium. An emanation was obtained by boiling the soil and 
 surface water at New Haven, which decayed at the same rate as 
 the radium emanation. 
 
 By exposing wires for several days in the open air, Bum- 
 stead 3 also observed that, after the excited activity due to the 
 radium emanation had disappeared, a part of the activity de- 
 cayed much more slowly. This residual activity decayed at the 
 same rate as the excited activity from thorium, showing conclu- 
 sively that the thorium as well as the radium emanation was 
 present in the air. Dadourian 4 showed that the soil at New 
 Haven was impregnated with the thorium emanation. A hole 
 was dug in the ground and the top closed. A negatively 
 charged wire was exposed in the hole, and on removal it was 
 found to show activity, which disappeared at the characteristic 
 rate of the excited activity of thorium. 
 
 Such results show that the soil at New Haven must contain 
 quite appreciable quantities both of thorium and radium. Since 
 the thorium emanation has a very short life, it is only able to 
 diffuse into the open air from a small depth of soil. The radium 
 
 1 Rutherford and Allan : Phil. Mag., Dec., 1902. 
 
 2 Bumstead and "Wheeler: Amer. Journ. Science, Feb., 1904. 
 
 3 Bumstead: Amer. Journ. Sci., July, 1904. 
 
 4 Dadourian: Amer. Journ. Sci., Jan., 1905. 
 
200 RADIOACTIVE TRANSFORMATIONS 
 
 emanation, which has a much longer life, is able to emerge from 
 a much greater depth. 
 
 In the meantime, C. T. R. Wilson 1 had found that rain was 
 radioactive. Rain water was collected after a shower, and 
 rapidly evaporated to dryness in a platinum dish, which was 
 then placed under an electroscope. The activity was found to 
 decay to half value in about 30 minutes. 
 
 Wilson in England, S. J. Allan and McLennan in Canada, 
 independently showed that freshly fallen snow possessed a like 
 property. The activity of snow, like that of rain, falls to half 
 value in 30 minutes. This rate of decay is nearly the same as 
 that observed for the excited activity of radium, several hours 
 after removal of the emanation. Such a result suggests that 
 the carriers of radium B and radium C become attached, prob- 
 ably by diffusion, to the water drops or snowflakes in their 
 passage through the air. On evaporation, the active matter 
 remains behind. A heavy fall of rain or a snowstorm must thus 
 act as a means of temporarily removing a proportion of the 
 radium B and C always present in the air. 
 
 Elster and Geitel found that the air in confined spaces, such 
 as caves and cellars, was abnormally radioactive, and showed 
 strong ionization. To show that these effects did not result 
 from stagnant air alone, Elster and Geitel confined a large 
 volume of air in an old steam boiler, but did not observe any 
 increase of the ionization with time. Other experiments showed 
 that the increased radioactivity in confined spaces, in contact 
 with the earth, was due to the gradual storing of the radium 
 emanation which diffused through the soil. In order to throw 
 light on this question, Elster and Geitel 2 placed a pipe several 
 feet deep in the earth, and by means of a pump sucked up some 
 of the air imprisoned in the capillaries of the soil. This was 
 found to be strongly active, and its activity decayed at about 
 the same rate as that of air mixed with the radium emanation. 
 
 Similar results were observed by Ebert and Ewers 3 for the 
 
 1 Wilson : Proc. Camb. Phil. Soc., xi, p. 428 (1902) ; xii, p. 17 (1903). 
 
 2 Elster and Geitel : Physik. Zeit., iii, p. 574 (1902). 
 8 Ebert and Ewers : Physik. Zeit., iv, p. 162 (1902). 
 
RADIOACTIVITY OF THE EARTH 201 
 
 air removed from the soil at Munich. Such results show con- 
 clusively that small quantities of radium are everywhere distrib- 
 uted throughout the surface soil of the earth. J. J. Thomson, 
 Adams, and others examined the water obtained from deep wells 
 and springs in England, and found that in some cases the water 
 contained considerable quantities of the radium emanation, and 
 in a few cases a trace of radium itself. 
 
 In the last few years, a very large amount of work has been 
 done in examining the waters and sediments of mineral and 
 hot springs for the presence of radioactive matter. H. S. Allan 
 and Lord Blythswood found that the hot springs at Bath and 
 Buxton contained appreciable quantities of a radioactive emana- 
 tion. This was confirmed by Strutt, who found that not only 
 was the radium emanation contained in the issuing water, but 
 that the mud deposited by the springs contained traces of 
 radium. It is of interest to note that helium has been observed 
 amongst the gases evolved by these springs, and it would appear 
 probable that the waters in their passage to the earth pass 
 through a deposit of radioactive minerals. 
 
 Himstedt found the radium emanation in the thermal springs 
 at Baden Baden, while Elster and Geitel found also small traces 
 of radium in the mud deposited by them. A large number of 
 springs have been examined by different observers in England, 
 Germany, France, Italy, and the United States, and in nearly 
 all cases the radium emanation has been found in the water, 
 often in easily measurable amount. Elster and Geitel found 
 that the mud or " fango " deposited from the hot springs at 
 Battaglia, Italy, was abnormally radioactive, and a close ex- 
 amination showed that the activity was due to radium. They 
 calculated that, weight for weight, it contains almost one thou- 
 sandth of the radium to be obtained from the Joachimsthal 
 pitchblende. 
 
 While the activity of most of the waters of hot springs is due 
 in most cases to the presence of radium or its emanation, Blanc l 
 has observed one notable exception in which the activity is mainly 
 due to thorium. The sediment of the waters at Salins-Moutiers 
 
 i Blanc : Phil. Mag., Jan., 1905. 
 
202 RADIOACTIVE TRANSFORMATIONS 
 
 was abnormally active, and was found to give off considerable 
 quantities of the thorium emanation. Blanc, however, was 
 unable to detect analytically the presence of thorium, although, 
 from the amount of emanation given off, a considerable quantity 
 should have been present. It seems not unlikely that the activ- 
 ity observed was due not to the primary substance thorium, but 
 to its product, radiothorium, which was discovered by Hahn 
 (see page 68). This would give rise to thorium X and the tho- 
 rium emanation, but would be present in too minute a quantity 
 to be determined chemically. 
 
 Elster and Geitel observed that natural carbonic acid obtained 
 from great depths of old volcanic soil contained the radium 
 emanation, while McLennan and Burton found considerable 
 quantities of radium emanation in the petroleum from a deep 
 well in Ontario, Canada. 
 
 In most cases where spring water comes from great depths, 
 and especially if the water is hot* radioactive matter is found to 
 be present in abnormal amount compared with that found in the 
 soil itself. Such a result is not unexpected, for water, and par- 
 ticularly hot water, would tend to dissolve traces of radioactive 
 matter in the strata through which it passes, and also to become 
 impregnated with the radium emanation. In special cases, it 
 may happen that the water has passed through a deposit of 
 radioactive minerals, and in such a case, a very strong activity 
 is to be expected. 
 
 Elster and Geitel have made an extensive examination of dif- 
 ferent soils for radioactivity, and have found traces of radioac- 
 tive matter in nearly every case. The activity is most marked 
 in clayey soils, and is apparently due in many cases to the pres- 
 ence of small traces of radium. The observations as a whole 
 show clearly that radioactive matter is extraordinarily diffused in 
 nature, and it is difficult to find any substance that does not 
 contain a minute trace of radium. It does not appear likely 
 that uranium or radium differ in this respect from the inactive 
 elements. 
 
 The presence of radium can be noted by the electric test, 
 where chemical analyses would fail to detect the presence of 
 
RADIOACTIVITY OF THE EARTH 203 
 
 rare inactive elements, although they may exist in considerably 
 greater quantity than the radium itself. On general grounds, 
 such a wide diffusion of radioactive matter is not surprising, for 
 the soil of the earth at any point should contain a fairly thorough 
 admixture of a great majority of the elements found in the earth, 
 the rare elements being present in only minute proportion. 
 
 There can be no reasonable doubt that the radioactive matter 
 observed in the atmosphere is mainly due to the emanation of 
 radium and its transformation products, and probably in some 
 localities to traces of the emanation of thorium and actinium. 
 The supply of radioactive matter in the atmosphere is kept up 
 mainly by the diffusion and transpiration of the emanations 
 through the soil, while no doubt a part is supplied by the action 
 of springs and by the release of imprisoned gases. 
 
 On account of its comparatively slow rate of change, it is to 
 be expected that the amount of the emanation of radium will pre- 
 dominate over the other emanations in the atmosphere, for the 
 short life of the emanations of actinium and thorium prevent 
 their reaching the surface from any appreciable depth. While it 
 is probable that the supply of emanation from the earth to the 
 atmosphere varies in different localities, the action of winds and 
 air currents generally should tend to distribute the emanation 
 from point to point and to make its distribution more uniform. 
 
 All observers have noted that the amount of excited activity 
 to be obtained under definite conditions from the atmosphere is 
 very variable, and often alters considerably during a single day. 
 Elster and Geitel made a detailed examination of the effect 
 of meteorological conditions on the amount of active matter in 
 the atmosphere. The experiments were made at Wolfenbiittel, 
 Germany, and were continued for twelve months. On an aver- 
 age, the amount of active matter increased with a lowering of 
 the temperature. Below C., the average was 1.44 times as 
 great as above C. A falling barometer increases the amount 
 of active matter. The effect of a change of pressure is intelli- 
 gible when it is remembered that a lowering of the pressure 
 tends to cause the emanation in the capillaries of the soil to be 
 drawn to the surface. 
 
204 RADIOACTIVE TRANSFORMATIONS 
 
 If the emanation observed in the atmosphere is entirely drawn 
 from the soil, the amount in the air over the sea in mid-ocean 
 should be much smaller than on land, for the water will not 
 allow the emanation to escape from the earth's crust into the 
 atmosphere. Observations so far made indicate that the amount 
 of active matter in the air falls off near the sea. For example, 
 the amount on the Baltic coast was found by Elster and Geitel 
 to be only about one third of that found inland, but no systematic 
 examination has yet been made of the amount of active matter 
 in the atmosphere at great distances from land. 
 
 AMOUNT OP THE RADIUM EMANATION IN THE 
 ATMOSPHERE 
 
 Most of the experiments on the amount of active matter in 
 the air have been qualitative in character, but it is obviously 
 important to obtain some idea of the amount of the radium 
 emanation present in the atmosphere. Since the amount in the 
 atmosphere is kept up by a constant supply of fresh emanation 
 from the earth, it is convenient to express the amount of radio- 
 active matter in the atmosphere in terms of the amount of freely 
 emanating radium bromide which is required to keep up the 
 supply. 
 
 Some interesting experiments in this direction have recently 
 been made by A. S. Eve 1 at Montreal. The radioactive state of 
 the air in the neighborhood of Montreal appears to be normal, 
 and the number of ions present per cubic centimetre of the out- 
 side air is about the same as that observed at different localities 
 in Europe. 
 
 Some experiments were first made in a large iron tank in the 
 Engineering Building of Me Gill University. This tank was 
 8.08 metres high and 1.52 metres square, with a total volume of 
 18.7 cubic metres. In order to determine the amount of ex- 
 cited activity to be obtained from the tank, a long insulated 
 wire was suspended in its centre, and kept at a constant poten- 
 tial of 10,000 volts for three hours. The wire was then rapidly 
 
 i A. S. Eve : Phil. Mag., July, 1905. 
 
RADIOACTIVITY OF THE EARTH 205 
 
 removed, and coiled round a frame attached to an electroscope. 
 The rate of fall of the gold leaves then served as a measure of 
 the amount of active matter deposited on the wire. 
 
 A similar experiment was then made in a small zinc cylinder 
 of volume 76 litres. The emanation derived from 2x1 0~ 4 milli- 
 grams of radium bromide was introduced into the cylinder and 
 mixed with the air. The excited activity was concentrated, as 
 before, on a negatively charged wire, and after removal was 
 measured with the same apparatus employed in the first experi- 
 ment. Knowing the rate of discharge of the electroscope for 
 the active deposit obtained from a known quantity of radium, a 
 comparison of the results obtained with the large tank at once 
 gave the amount of emanation present in it. In this way, it 
 was calculated that one cubic kilometre of the air, containing 
 the same amount of emanation per unit volume as that observed 
 in the large tank, was equivalent to the emanation supplied by 
 .49 grams of pure radium bromide. 
 
 The tank in these experiments was in free communication 
 with the open air, and the amount of the excited activity was 
 unaltered when the air from the room was forced through the 
 tank. It thus seems reasonable to suppose that the air within 
 the tank contained the same amount of emanation per unit 
 volume as the outside air. No radioactive matter had ever been 
 introduced into the building where the tank was used, and, as 
 we shall see later, the rate of production of ions per cubic centi- 
 metre of the tank was lower than that ever previously recorded. 
 
 In order to verify this point, however, experiments were 
 made on another large zinc cylinder placed on the College 
 Campus, with its ends in free communication with the air. The 
 active deposit was collected on a wire suspended along the axis 
 of the tank, and tested as before. The average amount, how- 
 ever, was found to be only from one third to one fourth of that 
 observed for an equal volume of the large tank. No adequate 
 cause could be assigned for this discrepancy between the results 
 for the experiments in the iron tank and in the zinc cylinder, 
 unless the charged wire is unable for some reason to collect all 
 the active deposit from the cylinder placed in the open air. 
 
206 RADIOACTIVE TRANSFORMATIONS 
 
 On certain assumptions we can form a rough estimate of the 
 amount of radium emanation existing in the atmosphere. Sup- 
 pose the emanation is uniformly distributed in a spherical layer 
 10 kilometres deep around the earth, and that the emanation per 
 cubic kilometre is uniform, and equal to that observed at Mon- 
 treal. The surface of the earth is about 5 x 10 8 square kilo- 
 metres, and the volume of the shell 10 kilometres thick is 5 x 10 9 
 cubic kilometres. Taking the estimated value, .49 grams per 
 cubic kilometre, found from experiments on the large tank, 
 the amount of emanation in the atmosphere corresponds to 
 2.5 x 10 9 grams or 2,460 tons of radium bromide. 
 
 Now, about three quarters of the earth is covered with water, 
 through which no emanation can escape to the surface. If 
 the emanation arises from the land alone, the amount is thus 
 reduced to about one quarter of this, or 610 tons. Taking the 
 value obtained from measurements in the cylinder in the outside 
 air, the amount is found to be about 170 tons. 
 
 Several observers have shown that the amount of excited 
 activity present in the air at high altitudes is equal to, if not 
 greater than, the amount observed on the plains. It thus ap- 
 pears probable that in supposing that the emanation is distrib- 
 uted to an average height of 10 kilometres in the atmosphere we 
 are well within the truth. Until a very complete radioactive 
 survey of the atmosphere has been made, such calculations are 
 of necessity somewhat uncertain, but they certainly serve to give 
 the right order of magnitude of the quantities involved. 
 
 Since the emanation is half transformed in about four days, it 
 cannot diffuse into the air from any great depth in the earth, so 
 that the main supply of the emanation must come from a super- 
 ficial layer of the earth not many metres in thickness. A part 
 of the supply is probably due to deep seated springs, which may 
 bring up the emanation from greater depths, but the amount so 
 supplied is probably small compared with that escaping directly 
 through the pores of the soil. 
 
 We thus arrive at the important conclusion that a very con- 
 siderable quantity of radium, measured by hundreds of tons, is 
 distributed over the earth within a few metres of its surface. 
 
RADIOACTIVITY OF THE EARTH 207 
 
 It is for the most part, however, distributed in such infinitesimal 
 quantities that its presence can be detected only by the aid of 
 the electric method. 
 
 Eve (loc. cit.) found that a wire about one millimetre in diam- 
 eter charged to 10,000 volts, and suspended about 20 feet above 
 the ground, was only able to collect the active deposit from the 
 air in a cylindrical volume of radius lying between 40 and 80 cms. 
 This collecting distance is small compared with that to be ex- 
 pected for such a high voltage, for the writer has shown that the 
 positively charged carriers of the active deposit of radium and 
 of thorium travel in an electric field at about the same velocity 
 as the ion, i. e., they move with a velocity of about 1.4 cms. per 
 second under a potential gradient of one volt per cm. It seems 
 probable that the carriers of the active deposit, which must 
 remain suspended a long time in the atmosphere, adhere to the 
 comparatively large dust nuclei always present in the air, and 
 consequently move very slowly in an electric field, so that the 
 carriers can only be drawn in from the immediate vicinity of the 
 charged wire. 
 
 THE PENETRATING RADIATION AT THE EARTH'S 
 SURFACE 
 
 Since the radium emanation is everywhere present in the sur- 
 face of the earth and in the atmosphere, its transformation prod- 
 uct radium C must give rise to 7 rays, and these rays must 
 come in all directions from the earth and atmosphere. The 
 presence of such a penetrating radiation at the earth's surface 
 was independently observed by McLennan l and H. L. Cooke, 2 
 in Canada. McLennan worked with a large vessel, and observed 
 that the ionization of the air inside diminished about 37 per 
 cent when the vessel was surrounded by a thickness of 25 cms. 
 of water. Cooke worked with a small brass electroscope of about 
 one litre capacity. The rate of discharge of the electroscope 
 fell about 30 per cent when completely surrounded by a lead 
 screen 5 cms. thick. No further diminution was observed by 
 
 1 McLennan : Phys. Rev., No. 4, 1903. 
 
 2 Cooke: Phil. Mag., Oct., 1903. 
 
208 RADIOACTIVE TRANSFORMATIONS 
 
 placing a ton of lead around the apparatus. The radiation is of 
 about the same penetrating power as the 7 rays from radium, 
 and can be observed in the open air as well as in a building. 
 By placing blocks of lead in different positions in regard to the 
 electroscope, it was found that the radiation came about equally 
 from all directions, and was the same during the day as at night. 
 Such results are to be expected if the penetrating rays come 
 equally from the radioactive matter distributed in the earth and 
 atmosphere. The magnitude of the ionizing effect due to the 
 penetrating rays is, however, much greater than that due to the 
 7 rays from the amount of radium emanation in the atmosphere 
 calculated by Eve. It seems not unlikely that these penetrat- 
 ing rays may be given out by matter in general as well as by 
 radioactive bodies. 
 
 ELECTRICAL STATE OF THE ATMOSPHERE 
 
 It has long been known, from observation of the potential 
 gradient in the atmosphere, that the upper layers of the atmos- 
 phere are generally positively charged in regard to the earth. 
 There is consequently an electric field always acting between the 
 earth and the upper atmosphere. Since there is a distribution 
 of ions in the lower regions of the atmosphere, there must con- 
 sequently be a steady movement of negative ions upwards and 
 of positive ions downwards towards the earth. Since the car- 
 riers of the active deposit of radium have a positive charge, they 
 must tend to be deposited on the surface of the earth. Each 
 blade of grass and each leaf must consequently be coated with 
 an invisible deposit of radioactive material. A hill or mountain 
 top tends to concentrate the earth's field at that point, and there 
 should be a greater amount of active matter deposited on its 
 surface than on an equal area on the plains. This is in agree- 
 ment with the observations of Elster and Geitel, who found 
 that the ionization of the air on a mountain top was greater than 
 on a lower level. 
 
 A large number of observations have been made of the rela- 
 tive number of ions in the air at various localities under differ- 
 ent meteorological conditions. Many experimenters have used 
 
RADIOACTIVITY OF THE EARTH 209 
 
 the " dissipation apparatus " constructed by Elster and Geitel. 
 This consists of an open wire gauze connected with an electro- 
 scope. The rate of discharge of the electroscope is separately 
 observed when charged positively and negatively. While this 
 apparatus has proved of value in preliminary work on the ioniza- 
 tion of the atmosphere, the results obtained are only comparative, 
 and do not readily lend themselves to quantitative calculations. 
 The effect of wind is very marked in such an apparatus, and the 
 rate of dissipation is always higher when a wind is blowing. 
 
 A very useful portable instrument for determining the actual 
 number of positive and negative ions per cubic centimetre of 
 the air has been devised by Ebert. 1 By means of a fan driven 
 by clockwork, a steady current of air is drawn between two con- 
 centric cylinders. The inner cylinder is insulated and connected 
 with a direct reading electroscope. The length of the cylinder 
 is so adjusted that all the ions in the air are drawn to the elec- 
 trodes in their passage through the cylinder. Knowing the 
 capacity of the instrument, the velocity of the current of air, 
 and the constants of the electroscope, the number of ions per c.c. 
 of the air can easily be deduced. When the inner cylinder is 
 charged positively, the rate of discharge of the electroscope is a 
 measure of the number of negative ions in the air, and vice versa. 
 
 Measurements by Eberts and others show that the actual 
 number of ions per c.c. of air is subject to considerable fluctua- 
 tions and is dependent on meteorological conditions. The num- 
 ber usually varies between five hundred and several thousands, 
 and the number of positive ions is nearly always greater than 
 the number of negative. 
 
 Schuster 2 observed that the number of ions per c.c. in the air 
 in Manchester varied between 2300 and 3700. These numbers 
 give the equilibrium number of ions in the air when the rate of 
 production of fresh ions is balanced by the rate of their recom- 
 bination. If n v n 2 are the number of positive and negative ions 
 respectively per c.c. of air, and q the rate of production per c.c. 
 
 1 Ebert: Physik. Zeit., ii, p. 662 (1901); Zeitschr. f. Luftschiff-fahrt, iv, Oct., 
 (1902). 
 
 2 Schuster : Proc. Manchester Phil. Soc., p. 488, No. 12, 1904. 
 
 14 
 
210 RADIOACTIVE TRANSFORMATIONS 
 
 per second, then q = a n^ n 2 , where a is the coefficient of recom- 
 bination of the ions. By a slight modification of the apparatus 
 of Ebert, Schuster was able to determine the value of a for the 
 air under the normal conditions of experiment, and deduced that 
 the value of q in Manchester varied between 12 and 39. 
 
 The apparatus of Ebert was designed to measure the number 
 of free ions in the air which have the same mobility as the 
 ions produced by X-rays or the radiations from active bodies. 
 The velocity of the ions produced in air have been directly 
 measured by Mache and von Schweidler. The positive ion 
 moves 1.02 cms. per second, and the negative 1.25 cm. per 
 second, for a potential gradient of one volt per cm. These 
 velocities are slightly slower than those observed for the ions 
 produced in dust-free air by X-rays or the rays from radioactive 
 substances. 
 
 In addition to these swiftly moving ions, Langevin l has 
 shown that a number of slowly moving ones are also present, 
 which travel too slowly in an electric field to be removed by the 
 electric field used in the apparatus of Ebert. These ions move 
 with about the same velocity as the ions observed in flame gases 
 some distance from the flame. By using much stronger fields, 
 Langevin has determined the number of these heavy ions pres- 
 ent in the air, and concludes that they are about forty times as 
 numerous as the swiftly moving ones. It is possible that these 
 slowly moving ions are formed by the deposition of water round 
 the ion to form a minute globule, or by the adherence of the ion 
 to the dust which is always present in the air. 
 
 Since there is undoubtedly a continuous production of ions 
 in the air near the earth, it is a matter of great importance to 
 determine the cause or causes of this ionization. The most 
 obvious cause is the presence of radioactive matter in the atmos- 
 phere. But is the amount present capable of producing the 
 ionization observed ? In order to throw light on this important 
 point, Eve (loc. cit.) made the following experiment. The large 
 iron tank, previously described on page 204, was used. An 
 insulated cylindrical electrode passed down the centre of the 
 
 1 Langevin : Comptes rendus, cxl, p. 232 (1905). 
 
RADIOACTIVITY OF THE EARTH 211 
 
 tank and was connected to an electroscope. The electrode was 
 charged to a sufficient potential to obtain the saturation current, 
 which is a measure of the total number of ions produced per 
 second. A wire charged to 10,000 volts was then suspended 
 in the tank and the active deposit collected from it for a definite 
 "time. The activity imparted to the wire was measured imme- 
 diately after removal with an electroscope. 
 
 An exactly similar set of experiments was then made with a 
 much smaller zinc cylinder, the air of which was artificially sup- 
 plied with the radium emanation obtained by blowing air through 
 a radium solution. The saturation current was measured, and 
 also the amount of the activity imparted to a central electrode 
 under the same conditions as in the large tank. If the ioniza- 
 tion in the large tank is due entirely to the presence of the 
 radium emanation in it, then the ratio of the saturation currents 
 in the two tanks should be equal to the ratio of the activities 
 imparted to the collecting wires under the same experimental 
 conditions. This must obviously be the case, since the saturation 
 current serves as a measure of the amount of emanation present, 
 and so also does the activity imparted to the collecting wire. 
 
 The ratio of the activity on the collecting wire in the iron 
 tank to that in the emanation cylinder was found to be about 14 
 per cent less than the ratio of the corresponding saturation cur- 
 rents. Considering the difficulty of such experiments, the agree- 
 ment is as close as could be expected, and indicates that the 
 greater part, if not all, of the ionization observed in the iron 
 tank was due to the presence of the radium emanation. 
 
 Since there was every reason to believe that the air in the tank 
 contained the same amount of emanation as the outside air, this 
 result indicates that the production of ions in the outside air 
 is mainly due to the radioactive matter contained in it. Be- 
 fore such a conclusion can be considered as established, experi- 
 ments of a similar character must be made in various localities. 
 We are, in any case, justified in assuming that the radioactive 
 matter in the air plays a very important part in the produc- 
 tion of ions observed in the atmosphere near the surface of the 
 earth. 
 
212 RADIOACTIVE TRANSFORMATIONS 
 
 It is of interest to record that Eve found the number of ions 
 produced per c.c. per second in the iron tank to be 9.8. This is 
 the smallest rate of production of ions yet recorded for a closed 
 vessel. Cooke observed a value as low as 20 for a well cleaned 
 brass electroscope of about one litre capacity. 
 
 If the radioactive matter in the air is the cause of its ioniza- 
 tion, there should be a constant proportion between the rate of 
 production of ions in the air and the excited activity on the col- 
 lecting wire. The data so far collected by various observers ap- 
 pear to contradict such a connection. It is doubtful, however, 
 whether the measurements actually supply the data required. 
 
 There seems to be no doubt that the recombination constant 
 of the ions depends greatly on meteorological conditions, and on 
 the freedom of the air from nuclei. The variation of this con- 
 stant affects the equilibrium number of ions in the air deter- 
 mined by the apparatus of Ebert. In a similar way, the 
 excited activity imparted to a charged wire in the open air 
 will probably depend upon atmospheric conditions, although the 
 amount of emanation present may not have been changed. 
 Before any definite conclusion can be reached, it will be neces- 
 sary to take all these factors into account. A large number 
 of observations have been made in Germany on the effect of 
 meteorological conditions on the amount of dissipation measured 
 by Elster and Geitel's apparatus. We have already mentioned 
 the effect of a rising or falling barometer, in producing well 
 marked variations in the amount of active matter in the air. 
 The relation between potential gradient and dissipation has been 
 studied by Gockel and Zolss. The latter finds that the poten- 
 tial gradient varies in a marked manner with the dissipation. 
 A high potential gradient is accompanied by a low value of the 
 dissipation, and vice versa. A similar relation between the poten- 
 tial gradient and the amount of ionization determined by Ebert's 
 apparatus has been observed by Simpson in Norway. Elster 
 and Geitel, and Zolss have shown that the dissipation increases 
 with the temperature. Simpson found that at Karasjoh in Nor- 
 way, the average between temperatures of 10 C. and 15 C. was 
 about six times as great as that between 40 C. and 20 C. 
 
RADIOACTIVITY OF THE EARTH 213 
 
 A very complete series of observations on the annual varia- 
 tion of the potential gradient, ionization, and dissipation, was 
 made by Simpson l at Karasjoh, in Norway, situated within the 
 Arctic circle, in latitude 69. These results are of special in- 
 terest, for between November 26 and January 18 the sun did 
 not rise above the horizon, while between May 20 and July 22 
 the sun did not fall below the horizon. 
 
 The absence of the sun's rays apparently had no marked 
 effect on the magnitude of the quantities measured. There was 
 on an average a steady rise of the potential gradient between 
 October and February, and a steady fall of the ionization during 
 the same period. Such results indicate that the sun's rays have 
 little if any direct effect on the ionization of the air. 
 
 It is impossible here to discuss the numerous speculations 
 that have been advanced to account for the presence of a strong 
 positive charge in the upper atmosphere. This positive charge 
 must be steadily supplied from some source, for otherwise it 
 would be rapidly discharged by the ionization currents between 
 the upper and lower atmosphere. Our knowledge of the elec- 
 trical state of the upper atmosphere is at present too imperfect 
 to enable us to determine whether this distribution of the charge 
 is due to an effect of radiations from the sun, as some have 
 supposed, or to a separation of the positive and negative ions 
 continuously produced in the atmosphere. 
 
 INTERNAL HEAT OF THE EARTH 
 
 The problem of the origin of the earth's internal heat has 
 been a subject of intermittent discussion for more than a cen- 
 tury. The most plausible and the generally accepted view is 
 that the earth was originally a very hot body, and in the course 
 of millions of years has cooled down to the present state. This 
 process of cooling is supposed to be still continuing, with the 
 result that the earth will ultimately lose its internal heat by 
 radiation into space. 
 
 On this theory, Lord Kelvin bases his well known deduction 
 of the age of the earth as a habitable planet. From observations 
 
 1 Simpson: Trans. Roy. Soc. Lond. A, p. 61, 1905. 
 
214 KADIOACTIVE TRANSFORMATIONS 
 
 of bores and mines, it has been found that the temperature of 
 the earth increases steadily from the surface downwards, and 
 on an average this temperature gradient is found to be about 
 1/50 F. per foot, or .00037 C. per cm. In order to obtain 
 an estimate of the maximum age of the earth on this theory, 
 Kelvin supposed that the earth was initially a molten mass. 
 By an application of Fourier's equation, it is possible to de- 
 duce the temperature gradient at the surface of the earth at 
 any time after the cooling began, provided the initial tempera- 
 ture and the average conductivity for heat of the materials of 
 the earth are known. Taking the most probable value of these 
 numbers, Kelvin in his original calculations found that the time 
 required for the earth to cool from the temperature of a molten 
 mass of rock to its present state was about 100 million years. 
 In later calculations, using improved data, this estimate has been 
 cut down to about 40 million years. 
 
 On this theory, life cannot have existed on the earth for 
 more than 40 million years. This period has been thought by 
 many geologists and biologists to be far too short to account for 
 the processes of organic and inorganic evolution, and for the 
 geologic changes observed in the earth, and such a serious cur- 
 tailment of the time at their disposal has given rise to much con- 
 troversy. On the theory on which Kelvin bases this calculation, 
 there can be little doubt of the probable correctness of this 
 estimate of the age of the earth, although the experimental data 
 on which the calculations were based are of necessity somewhat 
 imperfect. This theory, however, assumes that the earth is a 
 simple cooling body, and that there has been no generation of 
 heat from internal sources, for Lord Kelvin pointed out that the 
 possible heat developed by the earth's contraction or by ordi- 
 nary chemical combination is not sufficient to affect appreciably 
 the general argument. 
 
 The discovery of the radioactive bodies, which emit during 
 their transformation an amount of heat at least one million 
 times greater than that observed in ordinary chemical changes, 
 throws quite another light on this question. We have seen 
 that radioactive matter is everywhere distributed through the 
 
RADIOACTIVITY OF THE EARTH 215 
 
 surface of the earth and in the atmosphere, and that the amount 
 of radium existing close to the surface is of the order of several 
 hundred tons. 
 
 It is of interest to calculate how much radium must be uni- 
 formly distributed in the earth in order to compensate for the 
 present loss of heat from the earth by conduction to the surface. 
 The heat in gram calories per second lost by conduction to the 
 surface of the earth is given by 
 
 where R = radius of the earth, K the heat conductivity of the 
 earth in C. G. S. units, and T the temperature gradient. Let X 
 be the average amount of heat liberated per second per cubic 
 centimetre of the earth's volume, owing to the presence of 
 radioactive matter. If the heat, (?, supplied per second is equal 
 to that lost by conduction to the surface, then 
 
 X% TT E s = 4 TT R 2 K T, 
 
 -KT 
 orX=3 . 
 
 Taking the average value of K .004, the value taken by Lord 
 Kelvin, and T= .00037, then 
 
 X = 1 X 10~ 15 gram calories per second 
 = 2.2 x 10~ 7 gram calories per year. 
 
 Now one gram of radium in radioactive equilibrium emits 
 876,000 gram calories of heat per year. Consequently the pres- 
 ence of radium to the amount of 2.6 x 10~ 13 grams per c.c., or 
 4.6 x 10~ 14 grams per unit mass would compensate for the heat 
 lost by conduction. 
 
 In this calculation, the amount of radioactive matter present 
 has been expressed in terms of radium. There is no doubt that 
 uranium, thorium, and actinium are also present, but the heat- 
 ing effects of these are expressed in terms of radium. On this 
 view, the total heating effect of radioactive matter present in the 
 earth is equivalent to that of about 270 million tons of radium. 
 
216 RADIOACTIVE TRANSFORMATIONS 
 
 Such an estimate does not appear to be excessive when it is 
 remembered that there is undoubted evidence that several hun- 
 dred tons of radium are present in a thin shell at the earth's 
 surface. Taking the estimate of Eve that about 600 tons of 
 radium are required to keep up the supply of emanation in the 
 atmosphere over the land, it can be calculated that this comes 
 from a superficial layer of the earth, about 18 metres in depth. 
 This is based on the assumption that the radium in the earth 
 is distributed uniformly in the amount previously calculated. 
 Such a thickness is of the order of magnitude to be expected 
 from general considerations. 
 
 The experiments of Elster and Geitel have shown that radio- 
 active matter is found in rocks and in soils in about the 
 amount required by this theory. The heating effect of this 
 radioactive matter must undoubtedly be taken into account in 
 deductions based on the temperature gradient observed at the 
 earth's surface. If the calculated amount of radium were dis- 
 tributed uniformly in the earth, the temperature gradient would 
 remain constant as long as the supply of radioactive matter 
 remains unchanged. If the radioactive matter existed near the 
 surface of the earth in amounts greater than this mean value, 
 the temperature gradient would be correspondingly greater than 
 the observed value. 
 
 While the data on which these deductions are based is of 
 necessity somewhat meagre, the evidence so far obtained is 
 sufficiently strong to cast grave doubts on the validity of the 
 calculations of the age of the earth, based on the view that it is 
 a simple cooling body. The temperature gradient observed in 
 the earth to-day may have remained sensibly constant for mil- 
 lions of years in consequence of the steady generation of heat in 
 the earth. 
 
 It does not seem feasible on this theory of the maintenance of 
 the earth's heat to fix with any certainty the age of the earth. 
 The radium present in the earth is derived from the parent sub- 
 stance uranium, and on this theory uranium must exist in the 
 earth in the proportion of about one part in fifty million. This 
 proportion does not seem excessive from present data. The 
 
RADIOACTIVITY OF THE EARTH 217 
 
 life of uranium is about 1000 million years, so that if the inter- 
 nal heat of the earth were due entirely to uranium and radium 
 the temperature gradient 1000 million years ago would only be 
 about twice that observed to-day. 
 
 It has already been pointed out that some of the uranium 
 minerals are undoubtedly several hundred million years old, 
 and the evidence suggests that some of them are of still greater 
 antiquity. The evidence deduced from radioactive data alone 
 points very strongly to the conclusion that on the lowest esti- 
 mate, the earth is several hundred million years old. 
 
 The radioactive data do not of themselves enable us to decide 
 whether the earth was originally a very hot body or not. The 
 theory that the earth was originally a molten mass seems to 
 have been largely the outcome of an attempt to explain the 
 internal heat of the earth. Some geologists, notably Professor 
 Chamberlin of Chicago, have long upheld the view that the 
 geologic evidence by no means supports such a conclusion. It 
 is not possible here, however, to do more than mention this 
 interesting possibility. 
 
 RADIOACTIVITY OF ORDINARY MATTER 
 
 It is a matter of general experience that every physical prop- 
 erty discovered for one element has been found to be shared by 
 others in varying degrees. For example, the property of mag- 
 netism is most marked in iron, nickel, and cobalt, but every sub- 
 stance examined has been found to be either feebly magnetic or 
 diamagnetic. It might thus be expected on general principles 
 that the property of radioactivity which is so marked in a sub- 
 stance like radium would be shown by other substances. 
 
 A preliminary examination at once showed that if ordinary 
 matter was radioactive at all, it was only so in a minute degree, 
 but later work by McLennan, Strutt, Campbell, Wood, and 
 others, has shown that ordinary matter does possess the prop- 
 erty of ionizing the gas to a small extent. Campbell 1 in par- 
 ticular has carefully examined this question, and the evidence 
 obtained by him affords very strong proof that ordinary matter 
 
 1 Campbell: Phil. Mag., April, 1905 ; Feb., 1906. 
 
218 RADIOACTIVE TRANSFORMATIONS 
 
 does possess the property of emitting ionizing radiations, and 
 that each element emits radiations differing both in character 
 and intensity. 
 
 Experiments on this subject are very difficult, as the ioniza- 
 tion currents measured are extraordinarily minute. The effects 
 are very complicated as each substance emits a rays, and pene- 
 trating rays, and the latter in some cases give rise to a marked 
 secondary radiation. 
 
 Campbell concludes that the a rays emitted from lead have a 
 range of ionization in air of about 12.5 cms., while those from 
 aluminium have a range of only 6.5 cms. On an average the 
 a rays emitted from ordinary matter have a considerably greater 
 range in air than the a rays from radium. A sample of the lead 
 employed was dissolved in nitric acid and tested by the ema- 
 nation method for the presence of radium, but not the slightest 
 trace was observed. 
 
 It does not necessarily follow that these a particles are iden- 
 tical in mass with the a particles of radium. They may possibly . 
 be hydrogen atoms, for if the a particles from ordinary matter 
 were helium atoms we should expect, for example, to find helium 
 in lead. 
 
 If the expulsion of a particles be taken as evidence of atomic 
 disintegration, a simple calculation shows that the life of ordi- 
 nary matter is of the order of at least one thousand times that 
 of uranium, i. e. not less than 10 12 years. 
 
CHAPTER X 
 PROPERTIES OF THE a RAYS 
 
 IN the previous chapters we have shown how prominent 
 a role the a rays play in radioactive phenomena, as compared 
 with the more penetrating /3 and 7 rays. Not only are they 
 responsible for most of the ionization observed in the neighbor- 
 hood of radioactive matter, but they are also directly concerned 
 with the rapid emission of heat energy from these substances ; 
 in addition, they generally accompany the transformation of the 
 different types of radioactive matter, while the /3 and 7 rays are 
 emitted only in the case of -a few products. Finally, we have 
 seen that there is good reason to believe that the a particle is to 
 be identified with an atom of helium. 
 
 In this chapter, we shall outline in some detail the more im- 
 portant properties possessed by the a rays, and especially by the 
 a rays emitted by radium and its products. On account of 
 their great intensity, the a rays frQm radium have been more 
 easily studied than the corresponding rays from feebly radioac- 
 tive substances like uranium and thorium. At the same time, 
 the evidence so far obtained indicates that the a particles from 
 all the radioactive substances have the same mass, and differ for 
 each product only in their initial velocity of projection. 
 
 The a rays differ from the /3 and 7 rays in the ease with 
 which they are absorbed by matter and by the comparatively 
 large ionization they produce in the air near to a radioactive 
 body. By examining the effect of adding thin screens of metal- 
 lic foil over radioactive matter, it was found that the a rays 
 from the radioactive substances differed in penetrating power. 
 
 We shall see later that the a rays from radium are completely 
 cut off by a layer of aluminium foil of thickness .04 mms., or by 
 a layer of air of thickness 7 cms. The ionizing action of the a 
 rays is consequently confined within a short distance, while that 
 
220 RADIOACTIVE TRANSFORMATIONS 
 
 of the /3 rays extends for several metres, and that of the 7 rays 
 for several hundred metres. 
 
 The a rays were at first thought to be non-deflectable by a 
 magnetic field, for the application of a magnetic field sufficiently 
 strong to bend away completely the (3 rays had no appreciable 
 effect on the a rays. 
 
 In 1901, the writer began experiments by the electric method, 
 to see if the a rays could be deflected in a strong magnetic field, 
 but with the weak preparations of radium (activity 1000) then 
 available, the electric effects were too small to push the experi- 
 ments to the necessary limit. Later, in 1902, using a prepara- 
 tion of activity 19,000, the experiments l were successful, and 
 the a rays were found to be deflected in passing through both a 
 magnetic and an electric field. 
 
 The direction of the deflection was opposite to that for the @ 
 rays, and this indicated that the a rays consisted of a flight of 
 positively charged particles. By measurement of the amount of / 
 deflection of the rays in passing through magnetic and electric 
 fields of known strengths, the mass and velocity of the a par- 
 ticle were determined. The value of e/m the ratio of the 
 charge carried by an a particle to its mass was found to be 
 about 6 X 10 3 , while the maximum velocity v of the particle 
 was found to be 2.5 x 10 9 cms. per second. 
 
 Since the ratio e/m for the hydrogen atom is about 10 4 , this 
 result indicated that the a particle was atomic in size, and, 
 assuming that the a particle carried the same charge as a hydro- 
 gen atom, had a mass about twice that of the hydrogen atom. 
 It must be remembered that the amount of the deviation of the 
 rays in a given magnetic field is minute in comparison with that 
 of the /3 rays. For example, the swiftest a particle projected 
 from radium at right angles to a magnetic field of 10,000 C. G. S. 
 units describes the arc of a circle of 40 cms. radius. The swifts 
 est particle from radium, which is projected with 96 per cent 
 of the velocity of light, describes under similar conditions a 
 circle of about 5 mms. radius. 
 
 i Rutherford : Physik. Zeit., iv, p. 235 (1902) ; Phil. Mag., Feb., 1903. 
 
PRpPERTIES OF THE a RAYS 221 
 
 Becquerel 1 confirmed the magnetic deflection of the a rays 
 from radium by means of the photographic method, and also L 
 showed that the a rays from polonium have a similar property. 
 Using some pure radium bromide as a source of rays, Des Cou- 
 dres a measured the deflection of a pencil of a rays in a vacuum 
 after passing through a magnetic and electric field. He found 
 e/m to be 6.3 x 10 3 and the velocity to be 1.64 x 10 9 cms. per 
 second. The values of e/m obtained by Rutherford and Des 
 Coudres were in good agreement, but the velocities varied con- 
 siderably. In the experiments of Des Coudres the a rays were 
 passed through a screen of aluminium. It will be seen later 
 that this reduces the velocity of the a particles, and that the 
 correct velocity of the swiftest a particle from radium is about 
 2 x 10 9 cms. per second, or about 1/15 of the velocity of light. 
 
 In 1905, the question was again attacked by Mackenzie, 3 
 using pure radium bromide as a source of rays. A photographic 
 method was employed, in which the a rays fell on a glass plate 
 coated on its lower surface with zinc sulphide. A photographic 
 plate was placed on the upper side of the glass plate, and was 
 acted on by the light from the scintillations produced by the a 
 rays in the zinc sulphide screen immediately below it. The 
 deflection of the pencil of rays was observed as before after pass- 
 ing through a magnetic and electric field. The a rays were 
 found to be unequally deflected by a magnetic field, showing 
 that the a particles varied either in mass or velocity. This dis- 
 persion of the rays by a magnetic and electric field made it diffi- 
 cult to deduce the constants of the rays with accuracy. Taking 
 the mean value for the dispersion of the deflected pencils, he 
 found the value of e/m to be 4.6 x 10 3 , and the velocity of the a 
 particles to vary between 1.3 x 10 9 and 1.96 x 10 9 cms. per sec- 
 ond, on the assumption that the a particles all carry the same 
 charge and have the same mass. 
 
 The importance of an accurate determination of efm for 
 the a particle had long been recognized because of the light it 
 
 1 Becquerel: Comptes rendus, cxxxvi, pp. 199, 431 (1903). 
 
 2 Des Coudres: Physik. Zeit., iv, p. 483 (1903). 
 
 3 Mackenzie: Phil. Mag., Nov., 1905. 
 
222 RADIOACTIVE TRANSFORMATIONS 
 
 would throw on the question whether the a particle is an atom 
 of helium. In all the methods so far described, a thick layer of 
 radium in radioactive equilibrium was employed as a source of 
 rays. On the theory of absorption of the a rays, put forward by 
 Bragg and Kleeman, which will be discussed later, it was recog- 
 nized that the a rays emitted from a thick or a thin layer of 
 radium must consist of a particles moving at different speeds. 
 The use of a complex pencil of rays was open to very serious 
 objections, for it was impossible to know whether the rays most 
 deflected in a magnetic field corresponded to the most deflected 
 rays in an electric field or not. 
 
 The simplest method of accurately determining the value of 
 e/m is to use a homogeneous source of rays, i. e., to use a radio- 
 active substance in which all the a particles escape at the same 
 speed. The writer found that a wire made active by exposure 
 to the radium emanation completely satisfied these essentials. 
 The active matter, consisting of radium A, B, and C, is deposited 
 in an extremely thin film on a negatively charged wire exposed 
 in the presence of the emanation. After three hours' exposure, 
 the activity of the wire reaches a maximum value. After 
 removal, radium A, which has a three minute period, is rapidly 
 transformed, and has practically disappeared after fifteen min- 
 utes. The activity remaining is then entirely due to radium C. 
 The a particles from radium C are all expelled with identically 
 the same velocity, for there is no appreciable dispersion of the 
 rays in a magnetic field. The particles projected into the wire 
 are completed absorbed, and those which escape do not suffer in 
 velocity in their passage through the very thin film of intensely 
 active matter. 
 
 Using 10 to 20 milligrams of radium in solution, a wire one 
 centimetre long can be made extremely active by an arrange- 
 ment similar to that shown in Fig. 24, page 100. The wire pro- 
 duces a strong photographic impression on a plate brought 
 near it. The chief drawback to such a source of rays % is that 
 the intensity of the rays falls off rapidly, and two hours after 
 removal is only 14 per cent of the initial value. 
 
 The apparatus shown in Fig. 42 is very convenient for the 
 
PROPERTIES OF THE a RAYS 
 
 223 
 
 determination of the magnetic deflection of the rays. An active 
 wire is placed in a groove A. The rays pass through a narrow 
 slit B and fall on a small piece of photographic plate at C. The 
 apparatus is enclosed in a cylindrical vessel P, which can be 
 rapidly exhausted of air. The apparatus is placed between the 
 pole pieces of a large electromagnet, so that the magnetic field 
 is parallel to the direction of the wire and slit, and uniform 
 over the whole path of the rays. The electromagnet is excited 
 by a constant current, which is reversed 
 every ten minutes. On developing the 
 plate two well defined bands are observed 
 corresponding to the pencils of rays which 
 have been deflected equally on opposite 
 sides of the normal. 
 
 If p is the radius of curvature of the 
 circle described by the rays in a uniform 
 
 field of strength H, then Hp = , where 
 
 v is the velocity of the rays, e the charge 
 on the particle, and m its mass. 
 
 Let d = deflection of the rays from 
 the normal measured on the photographic 
 plate, 
 
 a = distance of plate from slit, 
 
 b = distance of slit from the source. 
 
 FIG. 42. 
 
 Apparatus for deter- 
 mining the amount of de- 
 flection of a pencil of a 
 rays in a strong magnetic 
 field. 
 
 Then by a property of the circle, if the deflection d is small 
 
 compared with a, 
 
 2pd = a (a + b). 
 Consequently, 
 
 mv Ha (a -f b) 
 
 = = - - 
 
 In the actual photographs, using the wire as a source of rays, 
 the traces of the pencil of rays stand out clearly with well 
 defined edges, so that the value 2d, the distance of the inside 
 edge of one band to the outside edge of the other, can easily 
 be measured. 
 
224 RADIOACTIVE TRANSFORMATIONS 
 
 The value of Hp for the a particles emitted from radium C 
 was found in this way to be 4.06 x 10 5 . In a field of 10,000 
 C. G. S. units, the a particle consequently describes a circle of 
 radius 40.6 cms. 
 
 RETARDATION OF THE VELOCITY OF THE a PARTICLE IN 
 
 PASSING THROUGH MATTER 
 
 It was found by the writer l that the velocity of the a par- 
 ticle diminishes in its passage through matter. This is most 
 simply shown by a slight modification of the 
 experimental arrangement, described above, which 
 has been used by Becquerel. By means of mica 
 plates, placed at right angles to the slit, the appa- 
 ratus is divided into two equal parts. One half 
 of the photographic plate is acted on by the rays 
 from the bare wire, and the other by the rays 
 which have passed through an absorbing screen 
 placed over the wire. 
 
 A drawing of a photograph obtained by this 
 method is shown in Fig. 43. The two upper 
 FIG. 43. bands A represent the traces of the pencil of rays 
 Retardation of obtained by reversal of the magnetic field for the 
 the velocity of fn)m the uncovered ha lf o f the wire . the 
 
 the a particles in J . ' 
 
 passing through lower bands B were obtained tor the rays from 
 matter. the wire, when covered with eight layers of alu- 
 
 minium foil, each of thickness about .00031 cms. 
 The apparatus was exhausted during the experiment, so that 
 the absorption of the rays in air is negligible. 
 
 The greater deflection of the pencil of rays which have passed 
 through the aluminium is clearly seen in Fig. 43. It will be 
 shown later that the value of e/m for the particles does not ' 
 change in consequence of their passage through matter. The 
 greater deflection of the rays is then due to a decrease of their 
 velocity after passing through the screen. This velocity is 
 inversely proportional to the distance between the centres of the 
 bands. 
 
 1 Rutherford : Phil. Mag., July, 1905 ; Jan. and April, 1906. 
 
PROPERTIES OF THE a RAYS 
 
 225 
 
 We have seen that the a particles from radium C are all pro- 
 jected initially at the same speed. The absence of dispersion 
 of the rays after passing through the screen shows that the 
 velocity of all the a particles is reduced by the same amount in 
 traversing the screen. 
 
 The following table gives the velocity of the a particles from 
 radium C after passing through successive layers of aluminium 
 foil, each of thickness about .00030 cms. The velocity is ex- 
 pressed in terms of F" , the velocity of the a particles from 
 radium C with an uncovered wire. 
 
 Number of layers of 
 aluminium foil. 
 
 Velocity of 
 a particles. 
 
 
 
 1.00 F 
 
 2 
 
 .94 ' 
 
 4 
 
 .87 < 
 
 6 
 
 .80 ' 
 
 8 
 
 .72 ' 
 
 10 
 
 .63 ' 
 
 12 
 
 .53 ' 
 
 14 
 
 .43 " 
 
 14.5 
 
 not measurable 
 
 There is a marked weakening of the photographic effect of the 
 rays after passing through 10 layers of foil. The photographic 
 impression is weak, but distinct with 13 layers, and, using very 
 active wires, can be observed with 14 layers. On account of 
 this falling off of the photographic effect of the rays, very active 
 wires are required to produce an appreciable darkening of the 
 plate through more than 12 layers of foil. The lowest velocity 
 of the a particle so far observed was about .4 FJ>, which cor- 
 responded to the velocity of the rays after passing through 14 
 layers of foil. The photographic action of the a rays steadily 
 diminishes with increase of the absorbing layer, but falls off 
 very rapidly for a thickness greater than 10 layers of aluminium. 
 The velocity of the a particle measured in this way is still 
 considerable when the photographic action has almost ceased. 
 Such a result suggests that there is a critical velocity of the a 
 particles, below which they are unable to affect appreciably a 
 
 photographic plate. 
 
 15 
 
226 RADIOACTIVE TRANSFORMATIONS 
 
 A similar abrupt falling off is observed in the ionizing and 
 phosphorescent effects of the rays. From observations on thin 
 layers of radium, Bragg found that the ionizing action of the 
 rays from radium C ceased comparatively abruptly after travers- 
 ing 7.06 cms. of air. A similar result was observed later by 
 McClung by using an active wire coated with radium C as a 
 source of rays. 
 
 In a similar way the writer found that the scintillations pro- 
 duced by the a rays on a screen of zinc sulphide ceased sud- 
 denly when the rays passed through 6.8 cms. of air. If layers 
 of aluminium foil are placed over the active wire, the range 
 of ionization and phosphorescence is diminished by a definite 
 amount for each layer. Each layer of foil of the thickness 
 used in the photographic experiment was equivalent in stopping 
 power to about . 50 cms. of air. A photographic^effect of the a 
 rays was just observable through 14 layers of aluminium. This 
 corresponds to 7.0 cms. of air, nearly the same range at which 
 the ionizing and phosphorescent effects vanish. The three char- 
 acteristic actions of the a rays thus cease together when the 
 rays have passed through a definite distance of air or a definite 
 thickness of an absorbing screen. Unless the velocity of the 
 a particle falls off with great rapidity at the end of its course 
 in air, it would appear as if there were a critical velocity: of 
 /the a particle below which it produced no appreciable ioniz- 
 ing, photographic, or scintillating effect. This property of the 
 a particles will be discussed in more detail later. In any 
 case the rapid falling off of these three actions of the a par- 
 ticle at the end of its range indicates that there is a close con- 
 nection between them. The photographic action of the a particle 
 falls off in the same rapid manner as the ionizing action, and it 
 seems reasonable to suppose that the effect of the rays on a pho- 
 tographic plate is the result of the ionization of the silver salts. 
 
 In a similar way, it is possible that the scintillations observed 
 in zinc sulphide are primarily caused by ionization of the sub- 
 stance, and that the scintillations may arise as a result of the 
 recombination of such ions. The brightness of the scintillations 
 certainly depends on the velocity of the a particle. If the effect 
 
PROPERTIES OF THE a RAYS 
 
 227 
 
 of the a rays on zinc sulphide is, as some have supposed, purely 
 mechanical, and the scintillations result from a cleavage of the 
 crystals, it is not easy to see why this effect should fall off 
 suddenly, although the energy possessed by the particle is still 
 considerable. 
 
 ELECTROSTATIC DEFLECTION OF THE a RAYS 
 
 In order to measure the deflection of the a rays from radium 
 C in an electric field, the arrangement shown in Fig. 44 was 
 adopted. 
 
 The rays from the active 
 wire W, after traversing a 
 thin mica plate in the base of 
 the brass vessel M, passed be- 
 tween two parallel insulated 
 plates A and B, about 4 cms. 
 high and 0.21 mm. apart. 
 The distance between the 
 plates was fixed by thin 
 strips of mica placed be- 
 tween them. The terminals 
 of a storage battery were con- 
 nected with A and B, so that 
 a strong electric field could 
 be produced between the 
 two plates. The pencil of 
 rays after emerging from the 
 plates fell on a photographic 
 plate P placed a definite dis- 
 tance above the plates. By 
 means of a mercury pump 
 
 the vessel was exhausted to a low vacuum. In their passage 
 between the charged plates, the a particles describe a parabolic 
 path, and after emergence travel in a straight line to the photo- 
 graphic plate. By reversing the electric field the deflection of 
 the pencil of rays is reversed. 
 
 In Fig. 45 A shows the natural width of the line on the plates 
 
 ','. I'P 
 
 
 
 \\ u 
 
 
 1.1 If 
 
 
 ll If 
 
 
 II if 
 
 
 1 
 
 i 
 ,'< 
 
 '! 
 
 I 
 
 I 
 i 
 
 M 
 
 /? 
 
 m 
 
 B 
 
 IJ 
 
 ___TT_ 
 
 FIG. 44. 
 
 Apparatus for determining the deflection 
 of a rays in passing through a strong 
 electric field. 
 
228 
 
 RADIOACTIVE TRANSFORMATIONS 
 
 when no electric field is acting ; B and C show the traces of 
 the deflected pencils of rays for potential differences between the 
 plates of 340 and 497 volts respectively. For small voltages the 
 natural width of the line is broadened ; for increased voltages 
 the single line breaks into two and the width of the lines steadily 
 narrows. Such an effect is to be expected theoretically. It can 
 be easily shown that if D is the distance between the extreme 
 edges of the deflected band for a potential difference E, 
 
 mv 2 
 
 SEl* 
 
 (B-df 
 
 where e is the charge on 
 the a particle, m its mass, 
 v its velocity, I the dis- 
 tance of the photographic 
 plate from the end of the 
 parallel plates, d the dis- 
 tance between the parallel 
 plates. This simple equa- 
 tion holds only if the field 
 is strong enough to deflect 
 the a particle through a 
 distance greater than d in 
 
 its passage through the electric field. For small values of the 
 field, a modified form of this equation must be used. 
 
 The decrease in velocity of the rays in passing through the 
 mica screen was separately determined. In most of the experi- 
 ments, the mica plate reduced the velocity of the a particles 
 from radium C by 24 per cent. 
 
 WYI W 
 
 From the magnetic deflection the value of - - is known,, 
 
 FIG. 45. 
 
 Electrostatic deflection of the a rays. The 
 bands are drawn to scale from the actual pho- 
 tographs. Magnification about 3 times. 
 
 m v* . 
 of is 
 
 while from the electrostatic deflection, the value 
 determined. 
 
 From these two equations the values of e/m and v can at once 
 be deduced. Examined in this way, it was found that : 1 
 
 1 Rutherford : Phvs. Review, Feb., 1906. 
 
PROPERTIES OF THE a RAYS 229 
 
 (1) The value of e/m was unaltered by the passage of the a 
 particles through matter ; 
 
 (2) The value of e/m was very nearly 5 X 10 3 ; 
 
 (3) The initial velocity of projection of the a particles from 
 radium C was 2 x 10 9 cms. per second. 
 
 A similar method was applied to determine the value of e/m 
 and the velocity of the a particles emitted from radium A and 
 radium F (radiotellurium). In both cases the value of e/m 
 was 5 x 10 3 within the limits of experimental error. The initial 
 velocity of the a particles from radium A was about 86 per cent V 
 of that of the a particle from radium C, while the velocity of * 
 the a particle from radiotellurium was about 80 per cent of that 
 of the a particle from radium C. The experiments on the ve- 
 locity and value of e/m for the a particles from radium itself 
 and from the emanation are not yet fully completed, but the 
 results so far obtained indicate that the value of e/m will be the 
 same as in the other cases. 
 
 Such results show conclusively that the a particles from 
 radium and its products have identical mass, but differ in the 
 initial velocities of their projection. The arguments in favor of 
 the view that the a particle consists of an atom of helium carry- 
 ing two ionic charges have already been discussed in detail on 
 pagelL84) 
 
 Dr. Hahn, working in the laboratory of the writer, has found 
 that the a rays emitted by thorium B are deflected both in a mag- " 
 netic and in an electric field. These rays have a velocity about 
 10 per cent greater than those from radium C, but have the 
 same value of e/m. In these experiments, the thorium B was 
 deposited on a thin negatively charged wire, by exposure to the 
 thorium emanation emitted by the very active preparation of 
 radio thorium separated by Hahn (see page 68). It was also 
 found that the range in air of the a particles expelled from 
 thorium B, determined both by the electrical and scintillation 
 methods, was about 8.6 cms., or about 1.6 cms. greater than that 
 for the a particles from radium C. 
 
 Since the mass of the a particle from thorium B and the 
 radium products is the same, it appears probable that the same 
 
230 RADIOACTIVE TRANSFORMATIONS 
 
 equality also holds for the a particles from the other thorium 
 products. The mass of the a particles from actinium has not 
 yet been measured, but there is every reason to believe that it 
 has the same value as for radium. On this view, the only com- 
 mon product of the different radioactive bodies is the a particle, 
 which, as we have seen, is a projected helium atom. 
 
 SCATTERING OF THE a RAYS 
 
 It is well known that a narrow incident beam of ft or cathode 
 rays is scattered in its passage through matter, so that the emer- 
 gent pencil of rays is no longer well defined. This scattering of 
 the ft rays increases as the velocity of the ft particles diminishes. 
 In a theoretical paper, Bragg l pointed out that this scattering 
 of the ft rays is to be expected. The ft particle in its passage 
 through the molecules of matter enters the electric field of the 
 atom, and its direction of motion is consequently changed. The 
 smaller the kinetic energy of the ft particle the greater will be 
 the deflection of the path of some of the rays. If a narrow pen- 
 cil of ft rays falls on an absorbing screen, a portion of the rays 
 will suffer so much deflection that the emerging beam will con- 
 sist of a much wider cone of rays. 
 
 On account of their much greater kinetic energy, it is theV 
 oretically to be expected that the a particles will suffer much 
 less deflection of their path in their passage through matter than 
 the ft rays. The a particles must move nearly in a straight v 
 line, and pass directly through the atoms or molecules of matter 
 in their path, without much change in their direction of motion. 
 This theoretical conclusion of Bragg is borne out by experi- 
 ment. The scattering of the a particles is very small compared 
 with that of the ft particles moving at the same speed, so that a 
 narrow pencil of a rays after traversing an absorbing screen is 
 still well defined after emergence. At the same time there is 
 undoubtedly a small scattering of the rays in their passage 
 through matter, which must be taken into account. 
 
 If the a rays pass through air, for example, the width of the 
 trace of a pencil of a rays on a photographic plate is always 
 1 Bragg : Phil. Mag., Dec., 1904. 
 
PROPERTIES OF THE a RAYS 231 
 
 broader than in a vacuum. In addition the edges of the bands 
 are not nearly so well denned in air as in a vacuum. This result 
 shows that some of the a particles have suffered a change of di- 
 rection of motion in their passage through the molecules of air. 
 
 The arrangement adopted to determine the retardation of the 
 velocity of the a particles in their passage through matter 
 (Fig. 43) is not complicated by the scattering of the rays, since 
 the absorbing screen is placed over the active wire between the 
 source and the slit. If, however, the absorbing screen is placed 
 over the slit, the scattering of the a particles is at once seen by 
 the broadening of the trace of the rays on the plate. A broad 
 diffuse impression is observed on the plate instead of the narrow 
 band with well defined edges, observed when the absorbing 
 screen is placed below the slit. The amount of scattering 
 increases with the thickness of the screen. When eleven layers 
 of aluminium foil were placed over the slit, an amount nearly 
 sufficient to cut off the ionizing and photographic effects of the 
 rays an examination of the photographic plate showed that 
 some of the rays had been deflected about 3 from the normal. 
 A part of the rays may have been deflected through a consider- 
 ably greater angle, but their photographic action was too small 
 to be detected. 
 
 We may thus conclude that the path of the a particles, espe- 
 cially when their velocity is reduced, is deflected to an appre- 
 ciable extent in passing through matter. The fact that the 
 direction of motion of an a particle possessing such great energy 
 of motion, can be changed in its passage through matter, shows 
 that there must exist a very strong electric field within the atom, 
 or in its immediate neighborhood. The change of direction of 
 3 in the direction of motion of the particles in passing through 
 a distance of .003 cms. of matter would require an average trans- 
 verse electric field over this distance corresponding to more than 
 20 million volts per centimetre. Such a result shows out clearly 
 that the atom must be the seat of very intense electrical forces 
 a deduction in harmony with the electronic theory of matter. 
 
 We have seen that the velocity of the a particles from radium 
 C lose their photographic action when their velocity falls to 
 
232 RADIOACTIVE TRANSFORMATIONS 
 
 J about 40 per cent of the initial value. On account of the com- 
 plications introduced by the scattering of the a particles in 
 passing through matter, it is difficult to decide with certainty 
 whether this " critical velocity" of the a particle, below which 
 it fails to produce its characteristic effects, is a real or only an 
 apparent property of the rays. Without discussing the evidence 
 in detail, I think there is undoubted proof that this critical 
 velocity of the a particle has a real existence. 
 
 PHOTOGRAPHIC EFFECTS OF THE a RAYS FKOM A THICK 
 LAYER OF RADIUM 
 
 Since the a particles emitted by radium and its products 
 decrease in velocity in their passage through matter, the radia- 
 tions emerging from a thick layer must consist of particles 
 moving at widely different speeds. This must obviously be the 
 case, since the a particles which come from some depth below 
 the radiating surface are retarded in their passage through the 
 radium itself. 
 
 A pencil of rays from radium is consequently complex, and if 
 a magnetic field is applied perpendicularly to the direction of the 
 rays, each particle will describe the arc of a circle, the radius of 
 which is directly proportional to the velocity of the particle. 
 
 This unequal deflection of the a particles in a magnetic field 
 
 ogives rise to a "magnetic^ spectrum," in which the natural 
 
 width of the trace is much increased. This dispersion of the 
 
 complex pencil of rays has been observed by Mackenzie 1 and 
 
 by the writer. 2 
 
 We have seen that the a particle has comparatively little 
 photographic effect when its velocity falls to about .6 V , where 
 V is the maximum velocity of the a particle from radium C. 
 Since the a particles from the latter have a greater speed than 
 the a particles from any other radium product, we might thus 
 expect to obtain a magnetic spectrum corresponding to a par- 
 ticles whose velocity lies between .6 V and V . In an actual 
 photograph of a deflected pencil of rays, the writer observed the 
 
 1 Mackenzie : Phil. Mag., Nov., 1905. 
 
 2 Rutherford: Phil. Mag., Jan., 1906. 
 
PROPERTIES OF THE a RAYS 
 
 233 
 
 presence of rays whose velocities lay between .67 V and .95 V ; 
 while Mackenzie, by the method of scintillations, observed the 
 presence of rays having velocities between .65 V and .98 V . 
 
 When we remember that the photographic action of the /3 and 
 7 rays from the radium prevents the detection of weak photo- 
 graphic effects produced by the a rays, the observations are seen 
 to be in good agreement with theory. 
 
 Becquerel 1 early observed an interesting peculiarity in the 
 deflection of a pencil of a rays from a thick layer of radium in 
 passing through a uniform magnetic field. A narrow vertical 
 pencil of rays fell on a photographic plate which was placed 
 at right angles to the slit and inclined at a small 
 angle with the vertical. By reversing the mag- 
 netic field, two fine diverging lines S P, S P', 
 were observed on the plate (see Fig. 46). The 
 distance between these lines at any point repre- 
 sents twice the deflection of the pencil of rays 
 from the normal at that point. By careful meas- 
 urement, Becquerel found that these two diverg- 
 ing lines were not accurately the arcs of a circle, 
 but that the radius of curvature of the path of 
 the rays increased with the distance from the 
 source. Becquerel thought that the a rays from 
 radium were homogeneous, and concluded from 
 this experiment that the value of e/m of the 
 particles progressively decreased in their passage 
 through air, in consequence of an increase of m by accretions 
 from the air. 
 
 Bragg, 2 however, showed that this peculiarity in the trace of 
 the rays could be simply explained without any assumption of 
 an alteration in the value of e/m,, by taking into account the com- 
 plexity of the pencil of rays. The experimental arrangement 
 is diagrammatically shown in Fig. 46. S P and S P' represent 
 the diverging traces of the rays on the photographic plate in a 
 uniform magnetic field after emerging from the slit S. Let us 
 
 1 Becquerel: Comptes rendus, cxxxvi, pp. 199, 431, 977, 1517 (1903). 
 
 2 Bragg: Phil. Mag., Dec., 1904; April, 1905. 
 
 FIG. 46. 
 
234 RADIOACTIVE TRANSFORMATIONS 
 
 consider, for example, the outside edge of the trace at a point A. 
 The photographic effect at this edge of the trace is due to the 
 particles of lowest velocity, which are just able to produce pho- 
 tographic action at A. Consider next a point B further removed 
 from the source. The a particles, which produce the edge of 
 the trace, have the same velocity as in the first case ; but since 
 they have had to travel through a distance B R of air instead of 
 A R, they must have initially started with a greater velocity. 
 This must evidently be the case, since the a particle is retarded 
 in its passage through air. The average velocity of these a par- 
 ticles along their path is consequently greater than in the first 
 case, and the outside edge will be deflected through a smaller dis- 
 tance than would be expected if the average velocity were the 
 same for the two paths A R, B R. This will cause the trace of 
 the rays to show evidence of steadily increasing radius of curva- 
 ture as we proceed from the source, a result in agreement with 
 the observations of Becquerel. 
 
 Quite a contrary effect is produced on the inside edge of the 
 trace, for this is produced by the swiftest a particles from 
 radium, viz., those emitted from radium C. Since the velocity 
 of these particles decreases in their passage through air, the in- 
 side edge will show evidence of decreasing radius of curvature. 
 This will -have the effect of contracting the natural width of the 
 trace. This effect is, however, small, and would tend to be 
 masked experimentally by the scattering of the rays in their 
 passage through air. 
 
 There is another paradoxical effect exhibited by a complex 
 /pencil of radium rays. Becquerel 1 showed that the outside 
 edge of the trace of rays obtained in a magnetic field is unal- 
 tered by placing absorbing screens over the radium. In the 
 case of a homogeneous source of a rays, we have seen that the 
 pencil of rays suffers a greater deflection after passing through 
 an absorbing screen. The absence of this effect in a complex 
 pencil of rays from radium led Becquerel to believe that the a 
 particles from it did not decrease in velocity in their passage 
 through matter. We have seen, however, that for each indi- 
 
 1 Becquerel: Comptes rendus, cxli, p. 485 (1905) ; cxlii, p. 365 (1906). 
 
PROPERTIES OF THE a RAYS 235 
 
 vidual product of radium, the a particles do suffer a retardation 
 of velocity under such conditions. 
 
 The explanation of this apparent paradox is simple. The 
 outside edge of the trace of the complex pencil of rays is due to 
 the lowest velocity a particles which are just able to produce an 
 appreciable photographic effect. The velocity of these a par- 
 ticles has been seen to be in the neighborhood of .6 V , where V 
 is the initial velocity of projection of the particles from radium 
 C. When an absorbing screen is placed over the radium all the 
 a particles suffer a retardation of velocity. The outside edge 
 of the trace of the rays is produced by a particles of the same 
 velocity as before ; not, however, by the same a particles, but by 
 another set, whose velocity has been diminished to this minimum 
 amount in their passage through the absorbing screens. 
 
 The absence of increased deflection of the pencil of rays by 
 the addition of absorbing screens is thus to be expected. These 
 anomalies in the behavior of a complex pencil of rays show how 
 necessary it is to use a source of homogeneous rays for the 
 investigation of the properties of the a rays. 
 
 An account of these peculiarities of a complex pencil of a 
 rays has been given in some detail, partly because of their great 
 interest, and partly because the explanation of the effects has 
 been a subject of some discussion. 
 
 ABSORPTION OF THE a RAYS 
 
 It was early recognized that the a particles were stopped in 
 their passage through a few centimetres of air or by a few 
 thicknesses of metal foil. On account of the weak ionization 
 produced by uranium and thorium, it was not at first possible 
 to work with narrow cones of a rays ; but experiments were 
 made with a large area of radioactive matter spread uniformly 
 over a plate. The saturation current was measured between 
 this plate and another plate placed parallel to it at a distance of 
 several centimetres. As successive layers of aluminium or other 
 metal foil were placed over the active matter, the ionization cur- 
 rent was found to fall off approximately according to an expo- 
 nential law with the thickness of the screen. A thick layer of 
 
236 RADIOACTIVE TRANSFORMATIONS 
 
 radioactive matter was generally employed, and in the case of 
 radium the exponential law appeared to hold fairly accurately 
 over a considerable range. 
 
 Some experiments on the absorption of the a rays from an 
 active preparation of polonium were made in a different way by 
 Mme. Curie. The rays from the polonium passed through a 
 hole in a metal plate covered with a wire gauze, and the ioni- 
 zation current was measured between this plate and a parallel 
 insulated plate placed 3 cms. above it. No appreciable current 
 was observed when the polonium was 4 cms. below the alumin- 
 ium window, but as this distance was diminished, the current 
 increased very rapidly, so that for a small variation of distance 
 there was a large alteration in the ionization current. 
 
 This rapid increase of the current indicated that the ioniz- 
 ing property of the a rays ceased suddenly after traversing a 
 definite distance in air. By adding a layer of foil over the 
 polonium this critical distance was diminished. 
 
 The observed fact that the ionization current between two 
 parallel plates varied approximately according to an exponential 
 law with the thickness of the absorbing screen, when thick 
 layers of radioactive matter were employed, tended to obscure 
 the true law of the absorption of the a rays ; for an exponential 
 law of absorption had been observed by Lenard for the cathode 
 rays, and also in some cases for the X-rays. In 1904, the ques- 
 tion was attacked by Bragg and Kleeman, 1 both on the theoreti- 
 cal and experimental side, and the interesting experiments made 
 by them have thrown a great deal of fresh light on the nature 
 of the a rays and on the laws of their absorption by matter. 
 
 In order to account for their experimental results, Bragg 
 formulated a very simple theory of the absorption of the a rays. 
 On this theory all the a particles from a thin layer of radioac- 
 tive matter of one kind were supposed to be projected with 
 equal velocities and to pass through a definite distance in air 
 before absorption. The velocity of the a particle decreased in 
 its passage through air in consequence of the expenditure of its 
 kinetic energy in ionizing the gas. As a first approximation, it 
 
 1 Bragg and Kleeman : Phil. Mag., Dec., 1904; Sept., 1905. 
 
UNIVER 
 
 PROPERTIES OF THE a RAYS 
 
 237 
 
 was supposed that the ionization produced by a single a particle 
 per centimetre of its path was constant for a certain range, and 
 then fell off very abruptly after the particle had traversed a defi- 
 nite distance of air. This " range " of the a particle varied for 
 each a ray product on account of the differences in the initial 
 velocities of the a particles expelled from the separate products. 
 If an absorbing screen were placed in the path of the rays, the 
 velocities of the a particles from a simple product were all 
 diminished in a definite ratio, and the range in air of the emerg- 
 ing particles was reduced by an amount proportional to the 
 thickness of the screen and its density as compared with air. 
 
 FIG. 47. 
 
 FIG. 48. 
 
 In a thick layer of radioactive matter, containing onlyooue 
 simple source of rays, the rays from the surface will have the 
 maximum range a. Those emerging from a depth d of the 
 radioactive matter of density p compared with air, will have a 
 range in air of a p d. With a thick layer of radioactive matter, 
 the a particles emitted into the gas will vary widely in velocity 
 and will have all ranges in air between zero and the maximum 
 range a. 
 
 Suppose that a narrow pencil of a rays from a simple type of 
 radioactive matter R (Fig. 47) passes into the ionization vessel 
 A B through a wire gauze A. If the layer of active matter of 
 one kind is so thin that the a rays are not appreciably retarded 
 in their passage normally through it, the ionization at different 
 
238 RADIOACTIVE TRANSFORMATIONS 
 
 distances from the source is expressed graphically in Fig. 48 by 
 the curve A P M. The ordinates represent distance from the 
 source of radiation and the abscissae the ionization produced 
 in the vessel The ionization commences suddenly at A and 
 reaches a maximum at P, when the rays pass to the upper plate 
 B of the ionization chamber and then remains constant till the 
 source is reached. 
 
 With a thick layer, however, the rays have all ranges in air 
 between the maximum and zero, and as the ionization vessel 
 approaches the source, more and more of these a particles pass 
 into it. The ionization curve is consequently then represented 
 by a straight line A P B. 
 
 Theoretically, in order to obtain such results a narrow cone 
 of rays and a shallow ionization vessel must be used. If the 
 ionization vessel includes the whole narrow cone of rays, at all 
 distances, the falling off of the intensity of the radiation accord- 
 ing to the inverse square law need not be taken into account. 
 
 The experiments of Bragg and Kleeman show that these theo- 
 retical conclusions are approximately realized in practice. 
 
 First, let us consider a thin layer of radioactive matter of one 
 kind. This was obtained by evaporating a small quantity of 
 radium bromide solution in a vessel. The emanation is driven 
 off and the active deposit is transformed in situ. After about 
 three hours the activity is due only to the a rays from the 
 radium itself. The ionization curve obtained by Bragg and 
 Kleeman is shown in Fig. 49, curve A. When the ionization 
 chamber is more than 3.5 cms. above the source only a slight 
 current is observed. At 3.5 cms. the current increases very 
 rapidly and reaches a maximum at 2.85 cms. It then slowly 
 falls off with decreasing distance. The maximum range of the 
 a rays from radium itself is consequently 3.5 cms. 
 
 The corresponding curve for radium C is shown in the same 
 figure, curve B. This was examined by McClung, 1 using the 
 methods employed by Bragg and Kleeman. The radium C was 
 deposited as an extremely thin film on a wire by exposure to 
 the radium emanation. The rays from radium C had a maxi- 
 
 1 McClung : Phil. Mag., Jan., 1906. 
 
PROPERTIES OF THE a RAYS 
 
 239 
 
 mum range of about 6.8 cms, and the ionization fell off in a very 
 similar way to that observed by Bragg for the radium rays. 
 
 In Bragg's experiment the ionization chamber had a depth of 
 2 mms., while in the experiments of McClung the depth was 
 5 mms. In the case of radium C the ionization is seen to be 
 nearly uniform for a distance of about 4 cms., and then to in- 
 crease rapidly, the maximum ionization being reached at a dis- 
 
 10 
 
 is* 
 
 1ONIZATIQN. 
 
 FIG. 49. 
 
 tance of 5.7 cms. Allowing for the fact that the ionization 
 chamber had a sensible depth, and that a fairly wide cone of 
 rays was employed, it can be shown that the ionization must 
 increase rapidly at a distance of 6.8 cms., but not quite so 
 rapidly as the simple theory supposes. 
 
 From a comparison of the diminution of velocity of the a par- 
 ticle from radium C in passing through aluminium, it can be 
 readily calculated that the velocity of the a particles at the 
 
240 
 
 RADIOACTIVE TRANSFORMATIONS 
 
 elbow of the curve is about .56 of the initial velocity of projec- 
 tion. At this velocity the a particde appears to be most efficient 
 as an ionizer. 
 
 Bragg and Kleeman have examined by this method the range 
 of the rays of the different a ray products present in radium in 
 radioactive equilibrium. A thin layer of radium was employed, 
 and the ionization curve is shown in Fig. 50. 
 
 .6 
 
 o a * 6 e 10 /a /* /6 19. 
 
 i ON/2 AT /OH. 
 
 FIG. 50. 
 
 The first a rays entered the testing vessel at a distance of 
 7.06 cms. from the source. These rays were emitted from 
 radium C, and have the greatest range of all the rays from the 
 radium products. At a point b the curve suddenly turns through 
 an angle, showing that at this point the a rays from another prod- 
 uct, whose range in air is 4.83 cms., have entered the testing ves- 
 sel. There is a similar though not so well defined break in the 
 
PROPERTIES OF THE a RAYS 241 
 
 curve at d for a distance 4.23 cms., showing that another set of 
 rays has entered the testing vessel. The break at/ is due to 
 the appearance of the rays from radium itself in the vessel. 
 We may conclude from these results that the a particles from 
 radium have a range of 3.5 cms. in air, and those from radium C 
 a range of 7.06 cms. The ranges 4.23 and 4. 83 cms. belong to 
 the emanation and to radium A, but on account of the rapid 
 change of A it has not yet been found possible to decide which 
 of these two numbers belongs to the rays from the emanation and 
 which to those from radium A. 
 
 If the curve o a I is produced downwards to c, the curve o a b c 
 represents the ionization due to radium C alone at different 
 distances from the source. Let this curve be now added to itself, 
 being first lowered through a distance 2.23 cms., corresponding 
 to the difference in range between 7.06 cms. and 4.83 cms. 
 The new curve b d e lies accurately along the experimental 
 curve b d. If the curve be again lowered through a distance 
 6.0 mms, corresponding to the difference of range for the next 
 products, and a similar addition be performed, the resulting 
 curve dfg again lies on the experimental curve. Finally if 
 the curve is lowered through 7.3 mms., it is similarly found 
 that the theoretical curve lies on the experimental curve/ h k. 
 
 Knowing the ionization curve of one product, the experi- 
 mental curve for the combined products caji thus be built up 
 from it in a very simple way. Such a result shows clearly that, 
 allowing for the differences in the initial velocities of projection, 
 the ionization curves for radium and each of its products are 
 identical. It also shows that the same number of a particles 
 are projected per second from each of the a ray products. This 
 result follows from the disintegration theory if the various 
 products are successive. 
 
 The results of Bragg and Kleeman have thus confirmed in a 
 novel and striking way the theory of successive changes, initially 
 developed from quite distinct considerations. They show that 
 the products are successive, for otherwise the experimental 
 curve could not be built up from consideration^ of the ioniza- 
 tion curve of one product alone. 
 
 16 
 
242 RADIOACTIVE TRANSFORMATIONS 
 
 We may thus conclude from this evidence that radium A and 
 C are true successive products, although it is difficult to test 
 this point satisfactorily by direct experiment. The results also 
 indicate that the a particles from all products are identical in 
 all respects except velocity a result confirmed, as we have 
 seen, by direct measurements. 
 
 The method developed by Bragg and Kleeman thus not only , 
 throws light on the nature of the absorption of a rays, but indi- J 
 rectly affords a powerful means of determining the number of a ] 
 ray products in radioactive matter, even if chemical methods 
 should fail to isolate these products from the parent substance. 
 This is possible if the a particles from the separate products 
 have different ranges in air. A series of breaks in the ioniza- 
 tion curve is a direct indication of the presence of a number 
 of distinct radioactive substances which emit a rays. By this 
 method, Dr. Hahn has shown that thorium B, which was sup- 
 posed to contain one product, in reality contains two. From 
 the difficulty of separation of these two products by chemical or 
 physical methods, it appears probable that one has an extremely 
 rapid period of transformation. 
 
 We have so far considered the ionization curves for a thin 
 layer of radium, as this brings out the essential features of the 
 absorption of a rays with great clearness. Bragg and Kleeman 
 have also determined the ionization curves for a thick layer of 
 radium. The curve is shown in Fig. 51. The curve consists 
 of a number of straight lines meeting at fairly sharp angles. 
 Above Q the ionization is due to the rays from radium C. At 
 Q the a particles from the product of range about 4.8 cms. 
 enter the ionization chamber, and the curve starts off at a sharp 
 angle. A similar break is observed at R and S when the a 
 particles from the other two products enter the ionization cham- 
 ber. The slopes of the curves P Q, Q R, R S, S T, are very 
 nearly in the ratio of 1, 2, 3, and 4, a result to be expected 
 from the simple theory. 
 
 Experiments were also made by Bragg and Kleeman on the 
 absorption of the a rays by thin metallic layers and by other 
 gases besides air. The effect of placing a uniform absorbing 
 
PROPERTIES OF THE a RAYS 
 
 248 
 
 screen over a thin layer of radioactive matter is to depress the 
 ionizatioii curve by the same amount throughout its whole 
 course. For example, the loss of range for an absorbing screen 
 of silver foil, whose weight per unit area was .00967 grams, was 
 equal to that for a stratum of air of thickness 3.35 cms. and 
 
 +5 
 .3.5 
 
 Pt.. 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 \ 
 
 \ 
 
 
 
 
 
 
 
 s 
 
 
 
 
 
 
 
 
 \ 
 
 N 
 
 
 
 
 
 
 
 x 
 
 > 
 
 
 3. f. 6 
 
 (P. 
 
 ?.. Q. 
 
 ~T. +j- <0. .. t- 
 
 fl^te o^ disc/icxrge in dr^itrary 5C<i/t 
 FIG. 61. 
 
 whose weight per unit area was .00402 gram. The ratio of 
 these two weights is 2.41, showing that the stopping power of 
 silver is 2.41 times greater than would be expected on a simple 
 density law. An examination of a number of metals showed 
 that the stopping power was approximately proportional to the 
 
244 RADIOACTIVE TRANSFORMATIONS 
 
 square root of the atomic weights. A similar law was found to 
 hold for gases over a considerable range of density. This rela- 
 tion is most remarkable, and indicates that the absorption of 
 energy in the atom is proportional to the square root of its 
 atomic weight. It is known that for simple gases like hydro- 
 gen, oxygen, and carbon dioxide, the total number of ions pro- 
 duced by complete absorption of a rays of given intensity is 
 nearly the same, suggesting that the same energy is required in 
 each case to produce an ion. If the stopping power of any gas 
 is mainly governed by the energy used up in producing ions, the 
 results obtained by Bragg and Kleernan indicate that on an 
 average four times as many ions are produced by the passage 
 of an a particle of given velocity through an atom of oxygen as 
 in its passage through an atom of hydrogen. This does not neces- 
 sarily assert that each atom of the gas in the path of the a par- 
 ticle is ionized, but is supposed to be an average result when a 
 larger number of atoms is considered. At the same time, there 
 is strong evidence that the number of ions produced by an a, 
 particle in air is at least as great as, if not greater than, the 
 number of molecules with which it collides. We are thus 
 driven to suppose either that the a particle is able to produce 
 more than two ions out of each molecule of a heavy gas, or that 
 the sphere of action of the a particle is greater in a dense than 
 in a light gas. 
 
 Whatever may be the conclusions to be drawn from the 
 experiments, they certainly serve to show that there is some 
 fundamental connection between ionization and the atomic 
 weight of different elements. 
 
 CHARGE CARRIED BY THE a RAYS 
 
 We have seen that the a particle is deflected in a magnetic or 
 electric field as if it carried a positive charge of electricity. It 
 was x early observed that the /3 particles from radium carried 
 with^hem a negative charge, and that the radium from which 
 they were expelled gained a positive charge. This property of 
 radium is illustrated in a striking manner in a simple apparatus 
 devised by Strutt, known as the "radium clock." Two gold 
 
PROPERTIES OF THE a RAYS 245 
 
 leaves are in metallic connection with an insulated tube con- 
 taining radium, and the whole is placed in a vessel exhausted to 
 a low vacuum. The /3 particles are fired through the radium 
 tube, carrying with them a negative charge, and leave behind an 
 equal positive charge. The leaves gradually diverge with posi- 
 tive electricity, and by a suitable contact are made to discharge 
 automatically after a certain divergence has been reached. This 
 process of charging and discharging continues indefinitely, or at 
 least as long as the radium itself will last. Using 30 milligrams 
 of radium bromide, the leaves may be made to pass through the 
 cycles of charge and discharge several times a minute. 
 
 If a rod or plate, covered with a thin film of radium, which 
 has been heated to get rid of the /3 and 7 rays, is exposed in a 
 similar way, no such charging action is observed, however good 
 a vacuum is produced. If the insulated plate is charged either 
 positively or negatively, the charge is rapidly lost. 
 
 Experiments of this kind are most simply made with a plate 
 coated with a thin film of radiotellurium (radium F).^ This sub- 
 stance has the advantage of emitting a rays but not ft rays. The 
 reason for the failure to detect the charge carried by the a rays 
 in the earlier experiments was made clear by an investigation 
 of J. J. Thomson. He showed that such an active plate emit- 
 ted in addition to the a particles a large number of slow mov- 
 ing electrons, which had very little power of penetration, and 
 moved with so slow a velocity that they could readily be turned 
 back or bent from their course by electric or magnetic fields. 
 The presence of a large number of these negatively charged 
 particles under ordinary conditions completely masks the charge 
 carried by the a particles. The effect of these slow moving 
 electrons, however, can be almost entirely eliminated by apply- 
 ing a strong magnetic field parallel to the plane of the active 
 plate. The electrons emitted from the plate then describe a 
 curved path in the magnetic field and return to the plate from 
 which they set out. Under such conditions, in a highly ex- 
 hausted vessel, it can be shown that the plate acquires a negative 
 charge, while a body on which the a particles impinge receives 
 a positive charge. 
 
246 RADIOACTIVE TRANSFORMATIONS 
 
 Such results show clearly that the a particles are expelled 
 with a positive charge, but that they are always accompanied by 
 a large number of slow moving electrons. These electrons ap- v 
 pear to be a type of secondary radiation set up by the escape of 
 the a particles from the active matter and by the matter on 
 which they impinge. Their presence has been noted not only 
 in radiotellurium but also in radium itself, in its emanation, and 
 in the emanation from thorium. These electrons appear to be 
 a necessary accompaniment of the emission of a particles, but 
 must not be confused with the /3 rays proper, which are pro- 
 jected at a much greater speed and have a much greater pene- 
 trating power. By employing a magnetic field to get rid of the 
 disturbance caused by the slow moving electrons, the writer 
 determined the charge carried by the a rays from a thin film of 
 radium spread uniformly on a metal plate. Knowing the quan- 
 tity of radium on the plate, it was deduced that 6.2 x 10 10 a 
 particles were projected per second from a gram of radium at its 
 minimum activity. For radium in equilibrium, which contains 
 four a ray products, the corresponding number is 2.5 x 10 11 . 
 These calculations are based on the assumption that each a par- 
 ticle carries a single ionic charge of value 3.4 x 10 10 electro- 
 static units. If the a particle carries double this charge the 
 number expelled is only one half of the above. 
 
 By measurements of the charge carried by the a rays from 
 radium C, obtained on a lead rod by exposure to the radium 
 emanation, it was calculated that 7.3 x 10 10 /3 particles are ex- 
 pelled per second from one gram of radium. Recently Schmidt 
 has shown that the supposed rayless product radium B, as well 
 as radium C, emits ft particles, but that these have much smaller 
 penetrating power. If equal numbers of /3 particles are ex- 
 pelled per second from radium B and C the number for each 
 ray product per gram of radium is 3.6 x 10 10 . 
 
 McClelland has shown that a strong secondary radiation is v 
 set up by the impact of /3 particles on lead. It is thus probable 
 that the number 3.6 x 10 10 is too high, for the /3 particles fired 
 into the lead give rise to secondary /3 particles whose charge is 
 measured with that of the primary /3 particles. If each a par- 
 
PROPERTIES OF THE a RAYS 247 
 
 tide carries twice the charge of the ft particle, the number of 
 particles expelled per second for each product in a gram of 
 radium should be 8.1 x 10 10 . Although it is difficult to draw 
 any very definite conclusions from such comparisons, the evi- 
 dence is in agreement with the view that in the product radium 
 C which emits a and /3 rays, the number of a and ft particles 
 emitted per second is the same, while the charge carried by the 
 a particle is twice that carried by the ft particle. 
 
 HEATING EFFECT OF THE a RAYS 
 
 In 1903, Curie and Laborde 1 made the striking discovery 
 that radium was always hotter than the surrounding medium, 
 and radiated heat at a constant rate of about 100 gram calories 
 per hour per gram. The question immediately arose as to 
 whether this phenomenon involved some new scientific prin- 
 ciple, or whether it was merely a secondary^effect due to the 
 bombardment of radium by its own a particles. 
 
 Since the a particles have large kinetic energy and are very 
 easily stopped by matter, most of those produced within the 
 radium do not emerge, but are stopped by the radium itself, and 
 their energy of motion is transformed into heat in situ. In 
 measurements of its heating effect the radium is enclosed in a 
 vessel of sufficient thickness to absorb all the a rays emitted 
 from the surface. It is consequently not necessary to make any 
 correction for the a particles which escape from the radium itself. 
 It thus appeared probable that the heating effect of radium 
 might result largely from the bombardment of the radium itself 
 by the a particles produced within it. 
 
 Rutherford and Barnes 2 made a number of experiments to 
 throw light on this subject. The heating effect of about 30 
 milligrams of radium bromide was first measured in a simple 
 form of air calorimeter and was found to correspond to about 100 
 gram calories per hour per gram. The radium was then heated 
 to a sufficient temperature to drive off the emanation, which 
 was then condensed in a small glass tube by means of liquid 
 
 1 Curie and Laborde : Comptes rendus, cxxxvi, p. 673 (1904). 
 
 2 Rutherford and Barnes : Phil. Mag., Feb., 1904. 
 
248 
 
 RADIOACTIVE TRANSFORMATIONS 
 
 air, and the tube sealed. The variation in the heating effect of 
 the radium so treated and of the emanation tube were then sepa- 
 rately examined. After removal of the emanation, the heating 
 effect of the radium fell rapidly in the course of about three 
 hours to 27 per cent of its maximum, and then slowly increased 
 again, finally reaching its old value after a month's interval. 
 
 The heating effect of the emanation tube varied in exactly 
 the opposite way, for it increased to a maximum in about three 
 
 In'it 
 
 I Heaf Ei fission 
 of Radi\ 
 
 / 
 
 o 9 
 <o 
 
 O 60 /6O 240 320 
 
 Xours 
 FIG. 52. 
 
 Variation of heating effect of radium after removal of its emanation and vari- 
 ation of that of the emanation and its products. 
 
 hours, when it was equal to about 73 per cent of the heating 
 effect of the radium. It then gradually decreased according to 
 an exponential law, falling to half value in about four days. 
 The curves of recovery of the heating effect of the radium and 
 of the loss of heating effect of the emanation tube are shown in 
 Fig. 52. Within the limit of experimental error, the sum of the 
 heating effects of the radium and emanation tube was equal to 
 that of the radium in radioactive equilibrium. Allowing for the 
 fact that 6 per cent of the emanation was not removed by the 
 
PROPERTIES OF THE a RAYS 
 
 249 
 
 heating, it is seen that only 23 per cent of the heating effect is 
 due to radium itself, and the other 77 per cent to the emanation 
 and its products. 
 
 The decay and recovery curves of the heating effect are iden- 
 tical within the limit of experimental error with the correspond- 
 ing curves of decay and recovery of activity measured by the a 
 rays. Such a result indicated that the heating effect is a meas- 
 ure of the kinetic energy of the a particles, for radium has an a 
 
 100; 
 
 \ 
 
 3,9 
 
 60, 
 
 UO M1NUTC4 
 
 FIG. 53 
 
 ray activity of about 25 per cent of its maximum when the 
 emanation and its products are removed, while the /3 and 7 ray 
 activity practically disappears. In order to test this point still 
 further, the distribution of the total heating effect of the emana- 
 tion tube between the emanation and its further products was 
 determined. After measuring the heating effect of the emana- 
 tion tube, the end of the tube was broken and the emanation 
 was entirely removed. Ten minutes later the heating effect of 
 the tube had dropped to 48 per cent, and it then diminished 
 
250 RADIOACTIVE TRANSFORMATIONS 
 
 steadily to zero. The curve of decrease of heating effect is 
 shown in Fig. 53. After ten minutes the curve follows the 
 activity decay curve fairly closely. After removal of the 
 emanation the heating effect of the tube is due to the active 
 deposit consisting of radium A, B, and C, which remains be- 
 hind. Since A loses its activity according to a three minute 
 period, it is not possible to follow the variation of its heating 
 effect. After fifteen minutes the heating effect must be due 
 entirely to radium B and C. It is not easy to decide experi- 
 mentally whether the rayless product radium B supplies an ap- 
 preciable proportion of the heating effect, but from the absence 
 of a rays, its heating effect is probably very small compared with 
 that of C. 
 
 The curve of increase of heating effect (Fig. 53) of a tube into 
 which the emanation has been introduced is complementary to 
 that of the curve of decrease. Such a relation is to be expected. 
 The fact that the heating effect falls off according to the period 
 of each a ray product shows that the heat emission of radium 
 and its products is mainly a consequence of the expulsion of a 
 rays. 
 
 It was deduced from the experiments that about 23 per cent 
 of the heating effect was due to radium alone, 32 per cent to 
 radium C, and 45 per cent to the emanation and radium A to- 
 gether. On account of the rapid rate of the change of A, its 
 heating effect cannot be disentangled from that of the emana- 
 tion. Direct experiment has shown that not more than one 
 or two per cent of the heat emission of radium is due to the 
 or 7 rays even when they are completely absorbed by a lead 
 envelope. 
 
 We shall now consider the important question as to whether 
 the energy of motion of the a particles ejected from radium and 
 its products is sufficient to account for the heating effect ob- 
 served. The kinetic energy of an a particle of mass m moving 
 with a velocity v is \mv*. The relative kinetic energies of 
 the a ray particles from each of the a ray products can at once 
 be determined if their relative velocities are known. These 
 velocities have not yet been directly measured for the rays from 
 
PROPERTIES OF THE a RAYS 251 
 
 each product, but they can readily be deduced from the ranges 
 of the a particles in air. For example, the velocity of the a 
 particle from radium itself is equal to that of the particle from 
 radium C after the latter has passed through a distance of air 
 equal to the difference of the ranges of the two a particles in air. 
 This difference is 3.5 cms., which corresponds to 6.7 layers of 
 aluminium foil of the same thickness as that employed in the 
 experimental results tabulated on page 225. In this way, taking 
 the kinetic energy of the a particle from radium C as 100, it 
 can readily be deduced that the kinetic energies of the a par- 
 ticles of range 4.8 and 4.3 cms. are 74 and 69 respectively. For 
 the a rays from radium itself of range 3.5 cms., the energy is 
 58. Since the products in radium are successive, the same num- 
 ber of a particles is expelled per second from each product. 
 Taking the kinetic energy of the a particle as a comparative 
 measure of its heating effect, it can at once be deduced from 
 these numbers that 19 per cent of the total heating effect should 
 be produced by radium itself, 48 per cent by radium A and the 
 emanation together, and 33 per cent by radium C. The corre- 
 sponding values obtained by direct measurement of the heating 
 effect are 23, 45, and 32 per cent respectively. The agreement 
 between theory and experiment is thus fairly good. 
 
 Now, on the assumption that the a particle carries an ionic 
 charge of 1.13 x 10~ 20 electromagnetic units, it has been experi- 
 mentally found that each of the a ray products present in one 
 gram of radium product expels 6.2 x 10 10 a particles per second. 
 The kinetic energy of the a particle from radium C is %mv 2 . 
 Substituting the known values e/m = 5 x 10 3 , and v = 2.0 x 10 9 , 
 ande = 1.13 X 10~ 20 , the kinetic energy is seen to be 4.5 x 10" 6 
 ergs. The value of the kinetic energy of radium C deduced 
 in this way is independent of whether the a particle carries one 
 or two ionic charges. The kinetic energy of the a particles of 
 radium C expelled per second from one gram of radium is thus 
 2.79 x 10 5 ergs. The heating effect of the a particles from 
 radium C per hour per gram of radium is consequently 24 gram 
 calories. The amount experimentally observed is 32 gram cal- 
 ories per hour. 
 
252 RADIOACTIVE TRANSFORMATIONS 
 
 Considering the experimental difficulty of accurately deter- 
 mining the number of a particles expelled from radium per sec- 
 ond, the agreement of experiment and calculation is remarkably 
 good, and shows clearly that the greater part of the heat emis- 
 sion of radium is due to self-bombardment by its own a par- 
 ticles. It is possible that a small fraction of the heat emitted 
 by radium may be due to the energy liberated in consequence of 
 the rearrangement of the atom after the violent expulsion of the 
 a particle, but it appears probable that this would be small com- 
 pared with the energy of motion of the a particle itself. 
 
 The conclusion that the heating effect of radium and its prod- 
 ucts is a measure of the energy of the a particles produced by 
 it applies also equally well to the other radioelements which 
 emit a rays. We should thus expect thorium, uranium, and 
 actinium to emit heat at a rate approximately proportional to 
 their a ray activities. Pegram has examined this question for 
 thorium and found evidence of a rate of heat emission about 
 that to be expected from its activity compared with that of 
 radium. Every substance that emits a particles must thus also 
 emit heat at a rate proportional to the product of the number of 
 a particles produced per second and the average kinetic energy 
 of each a particle. 
 
 The enormous heating effect of the radium emanation com- 
 pared with the quantity of matter involved has already been 
 discussed on page 91. The rapidly changing products like the 
 actinium, and. thorium emanations, and radium A, must initially 
 emit heat at an enormous rate compared with the quantity of 
 matter present. For example, the actinium emanation which 
 has a period of 3.9 seconds must, weight for weight, emit heat 
 at about 800,000 times the rate of the radium emanation. The 
 average duration of the time during which heat is emitted is 
 correspondingly less. 
 
 GASES EVOLVED FROM RADIUM 
 
 We have already seen that helium is produced in small 
 amounts from radium. Giesel, Runge, and Bodlander observed 
 that radium solutions produced a considerable quantity of hydro- 
 
PROPERTIES OF THE a RAYS 253 
 
 gen and oxygen. Ramsay and Soddy found that 50 milligrams f 
 of radium bromide in solution evolved about 0.5 c.c. of mixed < 
 gases per day. About 28.9 per cent of this consisted of oxygen 
 and the rest of hydrogen/TKe^ hydrogen is thus present in a 
 slight excess compared with that obtained by the decomposition 
 of water, and no satisfactory explanation of this excess has yet 
 been given. It may possibly be due to oxidation of the radium 
 bromide to radium bromate. Ramsay showed that the radium 
 emanation mixed with water produced hydrogen and oxygen, 
 and that after explosion of the mixed gases no observable 
 bubble of gas remained. This evolution of gas proceeds at a 
 constant rate, and must be the result of the action of the a 
 radiations in decomposing the water molecules. One gram of 
 radium bromide in equilibrium would produce about 10 c.c. of 
 hydrogen and oxygen per day. The energy required to disso- 
 ciate the corresponding amount of water per day is about 20 
 gram calories, or less than 2 per cent of the total kinetic energy 
 of the a particles produced by radium. 
 
 In order to generate 10 c.c. of hydrogen and oxygen per day 
 'by electrolysis a steady current of .00067 amperes is required. 
 Now it is found experimentally that the maximum ionization 
 current in air due to one gram of radium bromide in equilib- 
 rium, spread in a thin film, is .0013 amperes, or about twice 
 the current required to produce the amount of hydrogen and 
 oxygen observed. 
 
 In the experiments of Ramsay and Soddy some of the emana- 
 tion left the solution and collected in the open space above it, 
 and the rate of production of gases observed is probably less 
 than if the emanation were all retained in the solution. In 
 addition, the a rays not only decompose water but conversely 
 cause the combination of hydrogen and oxygen to form water. 
 Taking such factors into account, it certainly appears more* than 
 a coincidence that the ionization current in air due to the a 
 rays from radium is of the right magnitude to account for the 
 observed production of hydrogen and oxygen. 
 
 The gradual loss of energy of the a particles in their passage 
 through a gas appears to be mainly due to the energy absorbed 
 
254 RADIOACTIVE TRANSFORMATIONS 
 
 in the ionization of the gas. The fact that the stopping power 
 of matter in general, whether solid, liquid, or gaseous, is propor- 
 tional to the square root of the atomic weight, suggests that 
 matter of all kinds is ionized by the passage of the a rays. It is 
 thus to be expected that about the same total number of ions 
 would be produced by the complete absorption of the a rays in 
 water as are produced by absorption in air. The appearance of 
 hydrogen and oxygen in the radium solution is undoubtedly 
 mainly a result of the ionization of the water molecules by the 
 a particles, and shows that the ionization consists in large part 
 in an actual chemical dissociation of the water molecules. It 
 has been generally supposed that ionization in simple gases like 
 helium, hydrogen, and oxygen is due to the expulsion of an 
 electron from the molecule. This may be the case, but in a 
 complex molecule like that of water, the ionization by the a 
 rays consists in, or at any rate results in, an actual chemical dis- 
 sociation of water into hydrogen and oxygen. Whether this 
 dissociating action is a property only of the a rays, or of all 
 powerful ionizing agencies, cannot be discussed at this point, but 
 the evidence certainly suggests that the ionization of complex 
 substances by the a rays is very similar in character to ionization 
 of solutions, and consists in part in a chemical dissociation of 
 the substance. 
 
 There is considerable evidence that the a particles produce 
 chemical action of various kinds. For example, the a rays con- 
 vert oxygen into ozone, coagulate globulin, and produce chemi- 
 cal changes in barium platinocyanide. 
 
 SUMMARY OF PROPERTIES OF THE a RAYS 
 
 (1) The a particles from radium, and probably from all 
 radioactive substances, consist of positively charged atoms of 
 matter projected with great velocity. 
 
 (2) The a particles from radium and its products all have 
 the same mass, and are probably atoms of helium. 
 
 (3) Each product of radium expels a particles at a definite 
 speed, which is characteristic of that product, but varies for 
 different products. 
 
PROPERTIES OF THE a RAYS 255 
 
 (4) The ionizing, photographic, and phosphorescent actions 
 of the a rays from a simple product all appear to cease abruptly 
 when the velocity of the a particle falls below a certain critical 
 speed. 
 
 (5) The velocity of projection of the a particles increases 
 steadily for the successive products of radium, and is greatest for 
 those emitted from ^adium C. The maximum velocity is about 
 2 x 10 9 cms. per second. 
 
 (6) The velocity of the a particle is diminished in its passage 
 through matter. 
 
 (7) The a rays from a thin layer of any simple product are 
 homogeneous, i. ., they consist of a particles all projected with 
 the same velocity. On account of the retardation of the a par- 
 ticles in passing through matter, the rays from a thick layer of 
 any simple radioactive substance are complex, i. e., they consist 
 of a particles projected with velocities varying over a consider- 
 able range. 
 
 (8) The initial velocities of projection of the a particles from 
 radium and its products lies between 10 9 and 2 x 10 9 cms. per 
 second. 
 
 (9) The heating effect of radium is a result of the bombard- 
 ment of the radium by its own a particles. 
 
CHAPTER XI 
 PHYSICAL VIEW OF RADIOACTIVE PROCESSES 
 
 IN the preceding chapters, the more important properties of 
 the radioactive bodies have been considered, and it has been 
 seen that the results obtained receive a satisfactory explanation 
 on the view that the radioactive matter suffers spontaneous 
 disintegration. 
 
 We shall now endeavor to outline in as concrete a manner as 
 possible the processes which are believed to take place in the 
 atoms of radioactive matter, and in the medium surrounding 
 them. Such representations of the nature of the atom and of 
 the processes occurring therein are, in the present state of our 
 knowledge, somewhat speculative and imperfect, but they are 
 nevertheless of the greatest assistance to the investigator in 
 providing him with a working hypothesis of the structure of the 
 atom. The behavior of such model atoms can be compared with 
 that of the actual atoms of matter under investigation, and in 
 this way it is possible to form gradually a clearer and more 
 definite idea of the constitution of the atom. 
 
 Modern physical and chemical thories are all based on the 
 assumption that matter is discontinuous and is made up of a 
 number of discrete atoms. In each element the atoms are sup- 
 posed to be all of the same mass and of the same constitution, 
 but the atoms of the different elements show well marked dif- 
 ferences in physical and chemical behavior. It has been quite 
 incorrectly assumed by some that the study of radioactive phe- 
 nomena has tended to cast doubt on atomic theories. Far from 
 this being the case, such a study has materially strengthened, if, 
 indeed, it has not given actual proof of, the atomic structure of 
 matter. 
 
 Any one who has witnessed the multitude of scintillations 
 produced on a zinc sulphide screen by the a rays of radium can- 
 
RADIOACTIVE PROCESSES 257 
 
 not fail to have been impressed with the idea that radium is 
 sending out a shower of small particles. Such a view is con- 
 firmed by direct measurement, for we know that the scintilla- 
 tions are due to the a particles, which consist of minute 
 material bodies, all of the same mass, which are expelled from 
 the radium at an enormous speed. The energy of motion of 
 each a particle is so great that in some cases the impact of the a 
 particle on the screen is accompanied by a visible flash of light. 
 These a particles, as we have seen, are not fragments of radium, 
 but atoms of helium. 
 
 While the study of radioactivity has emphasized the ideas of i 
 the atomic constitution of matter, it has at the same time indi- \ 
 cated that the atom is not an indivisible unit but a complex 
 system of minute particles. In the case of the radioactive ele- 
 ments, some of the atoms become unstable and break up with 
 explosive violence, expelling in the process a portion of their 
 mass. Such views are rather an extension than a contradiction 
 of the usual chemical theory which supposes that the chemical 
 atom is the smallest combining unit of matter in ordinary chemi- 
 cal change. The atom may be the smallest combining unit, and 
 still be a complex system which cannot be broken up by any 
 physical or chemical agencies under our control. 
 
 In fact, the great emission of energy in radioactive changes 
 shows clearly the reason why chemistry has failed to break up 
 the atom. The forces which bind together the component parts 
 of an atom are so great that an enormous concentration of energy 
 would be required to break up the atom by the action of exter- 
 nal agencies. 
 
 The complex structure of the atom is clearly indicated by an 
 examination of its spectrum. Under the stimulus of heat or 
 the electrical discharge, the atom has certain definite periods of 
 vibration which are characteristic of each particular element. 
 Even in the case of a light atom like hydrogen, the great number 
 of different periods of vibration extending far into the ultra- 
 violet shows that the atom must be a complex structure which 
 is able to vibrate in a variety of ways. The modes of vibration 
 of a hydrogen atom are exactly the same under all conditions, 
 
 17 
 
258 RADIOACTIVE TRANSFORMATIONS 
 
 and are the same, for example, for free hydrogen in the sun and for 
 hydrogen prepared by different chemical processes on the earth. 
 The unchanging character of the spectrum of the elements 
 has been urged by some as an objection to the view that atoms 
 suffer disintegration. This objection, however, does not appear 
 to 'be a very weighty one, for the present theories of atomic dis- 
 integration suppose that there is not a gradual variation of the 
 properties of the atoms as a whole, but a sudden disintegration of 
 a minute fraction of the total number present, the rest remaining 
 quite unchanged. For example, the spectrum of radium itself 
 remains unaltered so long as any radium remains untransf ormed. 
 Supposing, however, we were able to detect the spectrum of the 
 products mixed with it, we should find the spectrum of radium 
 in equilibrium to consist of the normal radium spectrum, plus 
 the spectrum of each of its products superimposed upon it. 
 Each product would have a definite and characteristic spectrum, 
 differing for each product, and having no apparent connection 
 with that of the parent element. 
 
 DEVELOPMENT OF THE ELECTRONIC THEORY 
 OF MATTER 
 
 The laws of electrolysis discovered by Faraday indicated that 
 each atom of hydrogen carried an invariable charge, whose value 
 e could be approximately deduced from data on the mass of the 
 atom. The oxygen atom always carried a charge 2 e and an 
 atom of gold 3 e, and, generally, the ions of different elements in 
 solution carried charges which were integral multiples of the 
 charge carried by the hydrogen atom. In no case was any 
 atom found to carry a charge less than e. The idea thus arose 
 that the charge carried by the hydrogen atom was the smallest 
 unit of electricity, and was not capable of further subdivision. 
 Such a conception was practically equivalent to an atomic 
 theory of electricity. 
 
 Theories of atomic constitution were advanced in which it 
 was supposed that the atom consisted of a number of charged 
 ions in motion. Such theories, of which the most notable 
 exponents were Larmor and Lorentz, were primarily advanced 
 
RADIOACTIVE PROCESSES 259 
 
 to explain the mechanism of atomic radiation. More physical 
 definiteness was given to such theories by the discovery of J. J. 
 Thomson that the cathode rays consisted of a flight of particles, 
 whose apparent mass was only about 1/1000 that of the hydro- 
 gen atom. These " corpuscles " or " electrons " were found to 
 be emitted from substances under a variety of conditions. Not 
 only were they obtained by means of the electric discharge in a 
 vacuum tube, but also from a white hot carbon filament and 
 from a zinc or other metal plate exposed to the action of ultra- 
 violet light. They were found to be spontaneously emitted 
 from the radioactive bodies, with velocities in many cases much 
 greater than those attained in a vacuum tube. 
 
 At the same time the discovery by Zeeman of the effect of a 
 magnetic field on the period of the light vibrations indicated 
 that the vibrating system consisted of negatively charged par- 
 ticles, whose mass was about the same as that of the electron 
 set free in a vacuum tube. Such results indicated that the 
 electron was a constituent of all matter, and escaped from it 
 under a variety of conditions. 
 
 It was at first assumed as the simplest hypothesis that the 
 electron consisted of a material particle of mass about 1/1000 
 that of the hydrogen atom, and carrying the same charge as a 
 hydrogen atom in the electrolysis of water. Theory had long 
 before shown that a moving charge possessed electrical mass in 
 virtue of its motion. The theory indicated that this electrical 
 mass should be constant for slow speeds, but should increase 
 rapidly as the velocity of light is approached. In order to test 
 this theory definitely it was necessary to determine the value of 
 e/m for the electrons moving at velocities closely approaching 
 that of light. 
 
 Radium proved to be an ideal source of electrons for such an 
 experiment, since it expels a particles over a wide range of 
 speed, the velocity of some of them being very nearly equal to 
 that of light. As we have seen, Kaufmann measured the ve- 
 locity and value of e/m for the electrons from radium, and 
 definitely showed that the apparent mass of the electron in- 
 creased with its velocity. By comparison of theory with experi- 
 
260 RADIOACTIVE TRANSFORMATIONS 
 
 merit, it was found that the mass of the electron was purely 
 electrical in origin, and that there was no necessity to assume 
 that the charge was distributed over a material nucleus. We 
 thus arrive at the remarkable conclusion that the particles of the 
 cathode stream and the /3 particles of radium are not matter at 
 all in the ordinary sense, but disembodied electrical charges 
 whose motion confers on them the properties of ordinary mass. 
 It has already been pointed out (page 11) that ordinary mass 
 itself may possibly be explained purely as a result of electricity 
 in motion. In order to account for the observed magnitude 
 of the mass of the electron at different speeds, it is necessary 
 to suppose that the electricity is distributed over a surface of 
 minute area or throughout a minute volume. 
 
 Taking as the simplest assumption that this surface is spherical 
 in shape, it is necessary to suppose that the radius of the sphere 
 on which the charge is distributed is about 10~~ 13 cms. Now, 
 from a variety of considerations, it can be calculated that the 
 radius of an atom is about 10~ 8 cms., or rather that the sphere of 
 action of atomic forces extends from the centre of the atom over 
 about this distance. We thus see that if an atom were magni- 
 fied so as to be represented by a sphere of 100 metres radius, the 
 radius of an electron would be only one millimetre. Conse- 
 quently, if we suppose that the hydrogen atom consists of a 
 thousand electrons free to move within the dimensions of the 
 atom, the electrons would be so sparsely distributed that they 
 would occupy rather than fill the space within the atom, and 
 would only occasionally interfere with the freedom of one an- 
 other's individual motion. 
 
 Most of the magnetic and electric field surrounding an electron 
 in motion lies close to its surface. This must evidently be the 
 case, since the magnitudes of both these forces diminish in- 
 versely as the square of the distance, and become comparatively 
 small at a distance from the electron corresponding to a few 
 radii. The forces due to an electron in motion are thus mostly 
 confined within a sphere of radius about 10~ 12 cms. The direc- 
 tion or magnitude of the motion of any electron would not be 
 sensibly disturbed by the presence of another unless it ap- 
 proached within this small limiting distance. 
 
RADIOACTIVE PROCESSES 261 
 
 It has been found experimentally that an ion produced in 
 gases by X-rays, or the rays from active bodies, carries a positive 
 or negative charge equal in magnitude to 3.4 x 10~ 10 electro- 
 static units. The charge carried by an ion in gases is apparently 
 the same for all gases, and does not vary, as in the case of elec- 
 trolysis, with the valency of the atom. The charge carried by 
 an ion has been shown to be identical with that carried by a 
 hydrogen atom set free in the electrolysis of water. 
 
 Although the charge carried by an electron has not been 
 directly measured, there is every reason to believe that it is 
 identical with the charge carried by the negative ion in gases. 
 The charge on an electron is supposed to be the smallest 
 unit of electricity that takes part in the transfer of an elec- 
 tric current, whether in solids, liquids, or gases. There is one 
 marked point of distinction between a positive ion and an elec- 
 tron. The electron in motion has an apparent mass of about 
 1/1000 that of the hydrogen atom, while the corresponding 
 positive charge has never been found associated with a mass less 
 than that of the hydrogen atom. This has led to the view that 
 there is only one kind of electricity, viz., negative, which is 
 associated with the electron, and that a positively charged body 
 or ion is one which has been deprived of one or more of its 
 normal complement of electrons. 
 
 RADIATION FROM AN ELECTRON 
 
 An electron in motion produces a magnetic field whose in- 
 tensity at any point, for velocities small compared with that of 
 light, is proportional to its velocity. This magnetic field travels 
 with the electron, and magnetic energy is consequently stored 
 up in the medium surrounding it. The amount of this mag- 
 netic energy is proportional to the square of the velocity u of 
 the electron, and can consequently be expressed in the form 
 \ m u 2 . In this equation m represents the apparent or electrical 
 
 2 e 2 
 mass of the electron, and is equal to , where e is the charge 
 
 and a the radius of the electron. 
 
 An electron moving uniformly in a straight line does not 
 
262 RADIOACTIVE TRANSFORMATIONS 
 
 radiate energy, but any change in its motion is accompanied by 
 the dissipation of energy in the form of electromagnetic radia- 
 tion, which travels out from the electron with the velocity of 
 light. The rate of dissipation of radiant energy is proportional 
 to the square of its acceleration, and consequently becomes 
 large if an electron is suddenly set in motion or brought to rest. 
 For example, the X-rays produced in a vacuum tube are be- 
 lieved to consist of the intense electromagnetic pulses which 
 are set up by the sudden arrest of the cathode rays when they 
 impinge on the anticathode. 
 
 An electron constrained to move in a circular orbit is a power- 
 ful radiator of energy, since its motion is always accelerated 
 toward its centre. This necessary loss of energy from an accel- 
 erated electron has been one of the greatest difficulties met 
 with in endeavoring to deduce the constitution of a stable atom. 
 For the supposition that an atom consists of a number of posi- 
 itively and negatively charged particles in motion, held in 
 equilibrium by their mutual forces, Larmor 1 has shown that 
 the condition for no loss of energy by radiation is that the vector 
 sum of the accelerations of all the component charged particles 
 shall be permanently zero. If this condition is not fulfilled 
 there will be a steady drain of internal energy from the atom in 
 the form of electromagnetic radiation, and unless this is balanced 
 in some way by the absorption of energy from outside, the 
 atom must ultimately become unstable and break up into a new 
 system. 
 
 There are thus two essential conditions which must be ful- 
 filled for atoms to be permanent. The positively and negatively 
 charged particles constituting the atom must be so arranged as 
 to form a stable aggregate under their forces of attraction and 
 repulsion, and at the same time their arrangement and motions 
 must be such that no energy is radiated from the atom. 
 
 Since there is reason to believe that the atoms of many ele- 
 ments are either permanently stable or else remain stable for 
 intervals of time measured by millions of years, it would appear 
 that these two conditions must be very approximately realized in 
 
 1 Larmor : Aether and Matter, p. 233. 
 
RADIOACTIVE PROCESSES 263 
 
 the constitution of many of the atoms of the different elements. 
 Any atom in which these conditions were not fulfilled must 
 long ago have disappeared and been broken up into more stable 
 atomic systems. 
 
 It is thus not so much a matter of surprise that some atoms 
 spontaneously break up, as that the atoms are such stable 
 arrangements as they appear to be. The possibility of the 
 disintegration of the atom is thus in most cases a necessary 
 consequence of modern theories of atomic constitution. 
 
 REPRESENTATIONS OF ATOMIC CONSTITUTION 
 
 The recent developments in physical science have given a 
 great impetus to the study of the constitution of the atom, and 
 attempts have been made to form a mechanical, or rather elec- 
 trical, representation of an atom which shall imitate as closely as 
 possible the behavior of the actual atom. 
 
 On the electronic theory of matter it is supposed that the 
 hydrogen atom consists of about a thousand electrons held in 
 equilibrium by the internal forces of the atom. Since an atom 
 is electrically neutral in regard to external bodies, it is necessary 
 to assume that the negative charge carried by the electrons is 
 compensated by the presence within the atom of an equal posi- 
 tive charge. The electrons are supposed to be the mobile parts 
 of the atom, while the positive electricity is more or less fixed 
 in position. 
 
 The earliest representation of such a model atom was given 
 by Lord Kelvin. 1 A number of electrons or negatively charged 
 particles were supposed to be arranged within a uniform sphere 
 of positive electrification. The positive charge distributed 
 throughout the sphere was equal in magnitude to the corespond- 
 ing negative charges carried by the mobile electrons. This 
 arrangement was very ingenious, for it not only fulfilled the 
 condition that the atom was electrically neutral, but supplied 
 the necessary forces within the atom to hold the electrons in 
 equilibrium. 
 
 i Lord Kelvin : Phil. Mag., March, 1903; Oct., 1904 ; Dec., 1905. 
 
264 RADIOACTIVE TRANSFORMATIONS 
 
 Without such constraining forces it is obvious -that the elec- 
 trons would repel each other and escape from the atom. Lord 
 Kelvin showed that certain arrangements of the electrons 
 throughout the sphere were in stable equilibrium, while others 
 were unstable, and a slight disturbance would lead either to 
 their escape from the atom or to their falling into a more stable 
 configuration. Lord Kelvin has recently devised certain ar- 
 rangements of the positively and negatively charged particles 
 constituting the atom which are unstable, and must lead to the 
 expulsion of either a positively or negatively charged particle 
 with great velocity, thus imitating the behavior of a radioatom 
 in expelling a and /? particles. 
 
 The conception of an atom, suggested by Lord Kelvin, was 
 further developed by J. J. Thomson. 1 A number of electrons 
 arranged in a ring at definite angular intervals were supposed 
 to rotate uniformly within a sphere of positive electrification. 
 He drew attention to a remarkable property of such a config- 
 uration. We have seen that a single electron moving in a 
 circular orbit radiates energy, and the amount of the radia- 
 tion becomes large when the electron is supposed to describe 
 a circular orbit of atomic dimensions. When, however, a num- 
 ber of electrons followed one another in a circle, the fraction 
 of the energy of motion of the electrons radiated per revolu- 
 tion decreased very tepidly as the number of electrons in the 
 ring increased. 
 
 For example, the radiation from a group of six electrons mov- 
 ing with a velocity of one tenth that of light is less than one 
 millionth of that of a single particle. For a velocity of one 
 hundredth that of light the amount of radiation is only 10~ 16 of 
 that from a single electron moving with the same velocity in 
 the same orbit. 
 
 Such results show that an atom consisting of a number of 
 rotating electrons may radiate energy extremely slowly, but 
 ultimately this slow continuous drain of energy from the atom 
 results in a diminution of velocity of the electrons. When this 
 velocity falls below a certain critical value, the atom becomes 
 
 1 J. J. Thomson : Phil. Mag., Dec., 1903 ; March, 1904. 
 
RADIOACTIVE PROCESSES 265 
 
 unstable, and either breaks up with the expulsion of a part of 
 the atom, or forms a new arrangement of the electrons. 
 
 J. J. Thomson considers that the cause of the disintegration 
 of the atoms of radioactive matter must be ascribed to the loss 
 of energy by radiation from the atom. He has mathematically 
 investigated the possible temporarily stable arrangements of a 
 given number of electrons within a sphere of uniform positive 
 electrification. The properties of such an atom are very strik- 
 ing, and indirectly suggest a possible explanation of the periodic 
 law of arrangement of the elements in chemistry. When the 
 electrons revolve in one plane they tend to arrange themselves 
 in a number of concentric rings, and generally, if free to move 
 in any plane, in a number of concentric shells, like the coats of 
 an onion. 
 
 It is not necessary here to consider in detail the arrangements 
 discussed by J. J. Thomson, for these have already been given 
 by him in the Silliman Lectures of two years ago. It suffices 
 to say that such a model atom imitates in a remarkable way 
 the behavior of the atoms of the elements, and also suggests a 
 possible explanation of valency. 
 
 Some configurations of electrons, for example, are able to lose 
 one, others two or more electrons, and yet remain stable. Others 
 are able to gain an additional electron 6r two without altering 
 the main features of the arrangement of the electrons. Those 
 atoms which can readily lose electrons would correspond to an 
 electropositive element, and vice versa. 
 
 Such attempts to imitate by an electrical model the structure 
 of the atom are of necessity somewhat artificial, but they are of 
 great value as indicating the general method of attack of the 
 greatest problem that at present confronts the physicist. As 
 our knowledge of atomic properties increases in accuracy it may 
 yet be possible to deduce a structure of the atom which fulfils 
 the conditions required by experiment. A promising beginning 
 has already been made, and there is every hope that still further 
 advances will soon be made in the elucidation of the mystery of 
 atomic structure. 
 
 We have seen that on present theories positive electricity 
 
266 RADIOACTIVE TRANSFORMATIONS 
 
 plays a very different role from negative electricity. In order 
 to hold the electrons together and to make the atom electrically 
 neutral, it is necessary to call in the assistance of a fixed dis- 
 tribution of positive electrification. The mobile electrons con- 
 stitute, so to speak, the bricks of the atomic structure, while the 
 positive electricity acts as the necessary mortar to bind them 
 together. This appears to be a somewhat arbitrary arrangement, 
 but at present there appears to be no escape from this funda- 
 mental difficulty of the difference between positive and negative 
 electricity. 
 
 CAUSES OF ATOMIC DISINTEGRATION 
 
 We are now in a position to consider the possible causes that 
 lead to the disintegration of the atoms of the radioelements. It 
 has been shown that the law controlling the rate of disintegra- 
 tion of any individual product is very simple. The number of 
 atoms breaking up per second is always in a constant ratio to 
 the total number present. The value of this ratio, however, 
 varies enormously for the different products. It has not been 
 found possible to alter the rate of disintegration of any product 
 by any external agency. Difference of temperature, which plays 
 such an important part in altering the rate of chemical reac- 
 tions, is entirely without influence on the rate of transformation 
 of the radioactive bodies. For example, the heat emission of 
 radium, which is a measure of the kinetic energy of the a par- 
 ticles, is unaltered by plunging the radium into liquid hydrogen. 
 Elevation of temperature or chemical actions are equally without 
 influence. 
 
 It thus appears that the atoms of the radioelements suffer 
 spontaneous disintegration, or that it is brought about by forces 
 beyond our control. It has been suggested that the atoms of 
 radioactive matter may act as transformers of energy, abstracted 
 in some way from the surrounding medium. Theories of this 
 character were put forward for the special purpose of account- 
 ing for the emission of heat from radium without regard to the 
 nature of the other radioactive processes. There is indubutable 
 evidence that the heating effect of radium is a necessary conse- 
 
RADIOACTIVE PROCESSES 267 
 
 quence of the transformation of the radioatoms, and is a second- 
 ary effect resulting from the energy of motion of the expelled 
 a particles. 
 
 Such theories do not take into account the fact that radioac- 
 tivity is always accompanied by the appearance of new types of 
 active matter. There must consequently be chemical changes 
 in the active matter and, from other data, it is concluded that 
 the changes occur in the atom itself and not in the molecule. 
 
 The causes which lead to the disintegration of the atom are at 
 present a matter of conjecture. It is not yet possible to decide 
 with certainty whether the disintegration is due to an external 
 cause, or is an inherent property of the atom itself. It is con- 
 ceivable, for example, that some unknown external force may 
 supply the necessary disturbance to cause disintegration. In 
 such a case, the external force supplies the place of a detonator 
 to precipitate the atomic explosion. The energy liberated by the 
 explosion, however, is derived mainly from the atom itself and 
 not from the detonator. The law of transformation of radioactive 
 matter does not throw any light on the question, for such a law 
 is to be expected on either hypothesis. 
 
 It seems, however, most probable that the primary cause of 
 atomic disintegration must be looked for in the atom itself, and 
 consists in the loss of energy from the atom in the form of 
 electromagnetic radiation. We have seen that unless certain 
 conditions are fulfilled, an atom composed of negatively and 
 positively charged particles will lose energy by radiation and 
 ultimately break up. 
 
 For example, we have seen that J. J. Thomson has devised 
 certain models of atoms which radiate energy extremely slowly, 
 but which must ultimately, in consequence of the loss of atomic 
 energy, become unstable and either break up or form a new 
 atomic system. In the case of primary elements like uranium 
 and thorium, the atoms are comparatively stable and have an 
 average life of a thousand million years. The question then 
 arises whether this radiation of energy is continuously occur- 
 ring in all the atoms, or whether only a minute fraction is 
 involved at one time. On the first view, all atoms formed at 
 
268 RADIOACTIVE TRANSFORMATIONS 
 
 the same time should last for a definite interval. This, how- 
 ever, is contrary to the observed law of transformation, in which 
 the atoms theoretically have a life embracing all values from 
 zero to infinity. 
 
 We thus arrive at the conclusion that the configuration of 
 the atom which gives rise to a radiation of energy only occurs 
 in a minute fraction of the atoms present at one time, and is 
 probably governed purely by the laws of probability. 
 
 There is one peculiarity of the transformation of the products 
 of uranium, thorium, radium, and actinium which is possibly 
 important in this connection. The ft rays appear only in the 
 last of the rapid series of changes of these elements, and these 
 are expelled with enormous velocity. After the expulsion the 
 resulting product is either permanently stable or far more stable 
 than the preceding ones. It would appear more than a coin- 
 cidence that the expulsion of a high velocity /3 particle should 
 occur only in this final stage for each element. It is possible 
 that the particle, which is finally expelled, is the active agent 
 in promoting the previous transformations, and that when once 
 the disturbing factor has been removed, the resulting atom sinks 
 into a configuration of far more stable equilibrium. 
 
 For example, one of the electrons composing the atom may 
 take up a position in the atomic system which leads to a 
 radiation of energy. As a result the atom breaks up with the 
 expulsion of an a particle, and this process continues through 
 successive stages until, finally, there is a violent explosion 
 within the atom, which results in the expulson of the disturbing 
 electron with enormous speed. 
 
 PROCESSES OCCURRING IN RADIUM 
 
 Consider a minute quantity of radium of weight about one 
 millionth of a milligram in radioactive equilibrium. This will 
 contain about 3.6 x 10 12 atoms of radium of atomic weight 225. 
 Since in one gram of radium 6.2 x 10 10 a particles are expelled 
 per second from the radium itself, the number disintegrating per 
 second in one millionth of a milligram is 62. On an average, 
 an exactly equal number of a particles will be expelled from 
 
RADIOACTIVE PROCESSES 269 
 
 each of the successive a ray products, viz., the emanation, 
 radium A, and radium C. 
 
 The number of atoms of each of the radioactive products 
 present with this quantity of radium will be very different. 
 For 3.6 x 10 12 atoms of radium, there will be about 3 x 10 7 
 of atoms of emanation, 1.6 x 10 4 atoms of radium A, 1.5 x 10 5 
 of radium B, and 1.15 x 10 5 of radium C. The radium atoms 
 present will thus enormously preponderate over those of its 
 products. 
 
 Supposing it were possible to magnify this small particle of 
 radium so as to distinguish the individual atoms, we should see 
 a large number of radium atoms, and mixed with them a very 
 small number of atoms of its products ; but if the attention were 
 focussed on the atoms of each of the individual substances pres- 
 ent we should find that the same number of a particles are ex- 
 pelled from each of them per second. The number of atoms of 
 each product remains on an average constant, for the supply of 
 fresh atoms compensates for those that break up. 
 
 The electronic theory of matter supposes that an atom is 
 composed of a swarm of electrons in rapid movement held in 
 equilibrium by the internal forces of the atom. In the case of 
 heavy atoms, like those of the radioelements, it is not necessary 
 to suppose that each of these electrons has complete freedom of 
 movement. The character of the transformation of the atom 
 suggests that it is built up in part of a number of secondary 
 units, consisting of groups or aggregates of electrons in equilib- 
 rium, which are in rapid independent motion within the atom. 
 
 For example, it seems probable that the a particle or helium 
 atom actually exists as an independent unit of matter within 
 the radium atom, and is released at the moment of the disin- 
 tegration of the latter. These a particles"are in rapid movement, 
 and when a stage of instability is reached, one is ejected from 
 the atom with the velocity it possessed in its atomic orbit. If 
 this be the case, the a particles, on an average, must have a 
 velocity within the atom of more than 1/30 that of light. 
 
 It is possible, however, that a part of their great kinetic 
 energy may be acquired^ during the process of expulsion from 
 
270 RADIOACTIVE TRANSFORMATIONS 
 
 the atom, for we have seen that, from a variety of considera- 
 tions, the atom must be supposed to be the seat of intense elec- 
 trical forces. At the moment preceding the expulsion of an a 
 particle, from radium for example, the atom must be in a state 
 of violent disturbance. As a result, the forces which constrain 
 one of these a particles within the atom are momentarily neu- 
 tralized and the a particle escapes from the atom at an enormous 
 speed. 
 
 The internal forces are still sufficiently powerful to prevent 
 the escape of the other parts of the atom, and there is a rapid 
 adjustment of its components to form a new system which is 
 temporarily stable. It is probable that for a short interval 
 after the escape of the a particle, the atom is in a state of 
 violent disturbance, but finally sinks again into a temporarily 
 stable system. The residual atom has smaller mass than before, 
 and the internal arrangement of its parts is entirely different 
 from the previous one. The new atom, in fact, becomes an 
 atom of the emanation, and has chemical and physical properties 
 entirely different from those of the parent atom. 
 
 The atoms of the emanation are not nearly so stable as those 
 of radium, for they have an average life of only six days. The 
 atom breaks up as before, expelling another a particle and giv- 
 ing rise to an atom of radium A, which again differs widely in 
 properties from those of the emanation and of radium. This 
 substance is very unstable, for the atoms have an average life of 
 only four minutes. After the loss of another a particle the 
 atom of radium B makes its appearance. This atom undergoes 
 a change, which is apparently different in character from the 
 others. It may or may not expel an a particle, but if it does 
 so, the particle travels at too low a speed to produce appre- 
 ciable ionization in the gas. The atom, however, expels a /3 par- 
 ticle at a moderate speed. The atom of radium B then changes 
 into C. The instability of the latter atom results in an explo- 
 sion of great intensity. An a particle is expelled at a greater 
 speed than from any other product, while at the same time a /? 
 particle is ejected with a velocity nearly equal to that of light. 
 After this violent explosion the resulting atom sinks into a far 
 
RADIOACTIVE PROCESSES 271 
 
 more stable system, but this eventually breaks up, as we have 
 seen, and passes through still further stages. Finally, after the 
 expulsion of an a particle from radium F, the resulting atom is 
 probably identical with that of lead. 
 
 Such considerations show that a mass of radium is the seat of 
 an extraordinary conflict of forces. In a gram of radium, for 
 example, about 6 x 10 10 a particles are shot out each second 
 from each of the a ray products, while in addition there is an 
 expulsion of an equal number of high velocity electrons from 
 radium B and C. Since the a particles are only able to pass 
 through a small thickness of matter, the greater number of 
 these a particles are stopped in the radium itself, which is con- 
 sequently exposed to a bombardment of great intensity. 
 
 Let us concentrate our attention for a moment on an atom of 
 radium at the moment of the expulsion of an a particle. If the 
 principle of equivalence of momenta holds, the expulsion of the 
 a particle must cause a recoil of the atom from which it escapes. 
 Since the a particle has a mass of about 1/50 that of the radium 
 atom, and is expelled with a velocity of nearly 2 x 10 9 cms. per 
 second, the atom of radium must recoil with an initial velocity 
 of about 4 x 10 7 cms. per second, or about 200 miles per sec- 
 ond. This velocity will decrease rapidly in consequence of the 
 collisions of the moving atom with the atoms in its path, and it 
 will probably be brought nearly to rest after traversing a very 
 small distance. The kinetic energy of motion of the radium 
 atom will thus be transformed into heat. The a particle in- 
 itially starts with an enormous velocity, and must force its way 
 through the atoms of radium in its path, knocking off from 
 them a shower of electrons in the process. Its energy is gradu- 
 ally used up in producing ions, and its velocity consequently 
 diminishes. Finally, it loses "its power of ionizing, and is 
 brought to rest. Its charge is neutralized, and the a particle 
 then becomes a helium atom, and is mechanically imprisoned in 
 the mass of the radium. The energy used up in producing ions 
 in the radium is finally given up in the form of heat, for the 
 ions recombine, emitting heat during the process. 
 
 Let. us now turn our attention to the processes occurring in 
 
272 EADIOACTIVE TRANSFORMATIONS 
 
 the air or other gas surrounding the radium. The a particles 
 expelled from the surface of a mass of radium escape into the 
 air without loss of speed due to traversing the radium, and the 
 a particle from each product has its characteristic velocity. 
 Imagine that we can follow visually the flight of an a particle 
 through the gas. The velocity of the a particle is initially so 
 great compared with the velocity of translation of the molecules 
 of the gas, that the latter will appear to stand still during the 
 flight of the a particle. The time is too short for the molecules 
 of air to escape from the path of the a projectile, and its veloc- 
 ity and energy are so great that it is able to plunge through the 
 molecules in its path. The electric disturbance produced by its 
 passage through the molecule may lead to the expulsion of an 
 electron, and, probably in many cases, to the breaking up of the 
 complex molecule into charged atoms. 
 
 Two or more ions are consequently produced as a result of 
 the passage through the molecule. This process continues until, 
 after passing through about 3.5 cms. of air under normal condi- 
 tions, and producing about 100,000 ions, its power of ionization 
 is lost. Exactly what happens to the a particle at the end of its 
 career of ionization is not yet known. As we have seen, experi- 
 ment indicates that some of the a particles are still moving at a 
 high velocity when their ionizing power becomes very small. 
 Since the initial rapid reduction of the velocity of the a particle 
 appears to be mainly a result of the energy used up in ionizing 
 the gas, it is probable that the a particle, after its power of ioniz- 
 ing has been largely lost, will traverse a considerable distance of 
 air before it is brought to rest by continued collision with the 
 gas molecules. We are unable to detect the presence of such a 
 particles, since they have lost all the properties which serve 
 ordinarily to detect their presence. 
 
 There is a considerable amount of evidence to show that the 
 energy absorbed from the a particle in producing a pair of ions 
 is much greater than that required merely to separate the posi- 
 tive ion from the negative. Such a result suggests that during 
 the process of ionization the ions acquire a considerable velocity, 
 and that the energy spent in setting the ions in motion is large 
 
RADIOACTIVE PROCESSES 273 
 
 compared with that necessary for the mere separation of the 
 ions from the immediate sphere of each other's influence. 
 
 Although the ft particle escapes -from the radium with an 
 average velocity ten times that of the .a particle, it is far less 
 efficient as an ionizer. It produces only a. small number of ions 
 per centimeter of path compared with the number produced by 
 an a particle, and passes through about 100 times the distance in 
 air before it ceases to ionize. 
 
 We have not so far considered the connection of the 7 rays 
 with radioactive changes. These rays always appear with the 
 rays and are believed to be an electromagnetic pulse, set up in 
 consequence of the sudden expulsion of the ft particle. This 
 pulse is the seat of very intense electric and magnetic forces 
 and travels out from the atom like a spherical wave with the 
 velocity of light. Such a pulse is a very inefficient ionizer com- 
 pared with the a particle, and on an average produces only one 
 ion per centimeter of its path for 10,000 produced by the a par- 
 ticle. The penetrating power of the 7 ray, on the other hand, is 
 very great, and it continues to produce ions even after travers- 
 ing a great distance of air. 
 
 The energy of the rays which traverse the gas is ultimately 
 frittered down into heat. The initial energy of motion of the 
 ions is rapidly lost by collision with the gas particles, while the 
 ions finally recombine with the liberation of energy. 
 
 In addition to the ionization effects already considered, there 
 are very marked secondary effects produced when the rays 
 impinge upon matter. The a particles release a shower of 
 electrons from the matter upon which they impinge. These 
 electrons, however, are emitted at a very slow speed. On the 
 other hand, the ft and 7 rays cause the release of electrons at a 
 speed comparable with that of light. These secondary radia- 
 tions are most marked when the radiations fall on heavy metals 
 like lead, but no doubt occur, though in a much less intense 
 degree, during the passage of the rays through a gas. 
 
 On account of the great energy of motion of the a particle, it 
 might be expected to set in vibration the atoms of matter in its 
 path, and cause them to emit light waves. This property of the 
 
 18 
 
274 RADIOACTIVE TRANSFORMATIONS 
 
 a rays of exciting luminosity was first noted by Sir William and 
 Lady Huggins, 1 who found that the weak phosphorescent light 
 of radium showed the band spectrum of nitrogen. This has 
 been traced to the action of the a rays, either in free nitrogen 
 close to the radium, or in nitrogen occluded within the radium 
 compound. Such a result is of unusual interest, as it is the 
 first example of a gas giving a spectrum when cold without the 
 stimulus of a strong electric discharge. 
 
 Walter and Pohl 2 have recently found that the gas traversed 
 by the rays from an active preparation of radiotellurium emits 
 light waves which act on a photographic plate. The intensity 
 of this effect is greatest for pure nitrogen. The atoms of nitro- 
 gen appear to be more easily stimulated to give out their char- 
 acteristic vibrations than any other gas so far examined. It is 
 a matter of surprise that as yet no evidence has been obtained 
 that the a particles themselves give a spectrum. The violent 
 collisions of the a particle with the molecules in its path must 
 set up vibrations in the a particle, and it should give a charac- 
 teristic spectrum. Experiments of this character, though of 
 great difficulty, are most important, for they may throw light on 
 the nature of the a particle. In this connection it is of interest 
 to note that Giesel found that a preparation of "emanium" 
 emitted a phosphorescent light consisting of bright lines. These 
 lines were found to be due to didymium, which was present as 
 an impurity in the active substance. 
 
 There is no doubt that the radiations emitted from active 
 bodies serve as very powerful agents for the ionization and dis- 
 sociation of matter. No definite evidence has so far been ob- 
 tained that the a or /3 particles emitted from radium are able to 
 hasten its rate of transformation. Such a concentrated source 
 of energy as these high velocity particles might be expected to 
 produce, under some conditions, a disintegration of the atoms of 
 matter through which they pass. A mass of radium, for exam- 
 ple, which is subjected to an intense bombardment by its own a 
 
 1 Sir William and Lady Huggins: Proc. Roy. Soc., Ixxii, pp. 196, 409 (1903) ; 
 Ixxvii, p. 130 (1906). 
 
 2 Walter and Pohl: Ann. d. Phys., xviii, p. 406 (1905). 
 
RADIOACTIVE PROCESSES 275 
 
 and /3 particles might be expected to disintegrate faster than the 
 same amount of radium diffused through a large volume. Fur- 
 ther experiments in this direction may yet show that such an 
 effect does exist, but it is certainly not very marked. A direct 
 attack on the question as to whether X-rays are able to cause 
 the disintegration of matter has been made by Bumstead. 1 A 
 strong beam of X-rays fell on two plates of zinc and lead, 
 which were of such a thickness as to absorb an equal fraction of 
 energy of the incident beam. The lead was raised to a consider- 
 ably higher temperature than the zinc, indicating that although 
 the same fraction of the energy of the rays had been absorbed, 
 more energy had been released in the lead than in the zinc. 
 Such a result suggests that the X-rays cause a greater amount 
 of atomic disintegration in lead than in zinc, and that a consider- 
 able portion of the heat generated in the lead is due to the 
 energy liberated from the transformation of its atoms. 
 
 Further experiments with a variety of metals and with differ- 
 ent sources of intense ionizing radiations are required to sub- 
 stantiate completely such a far-reaching conclusion, but the 
 results so far obtained in this difficult field of research certainly 
 lead us to hope that we may yet bring about the disintegration 
 of atoms by laboratory methods. 
 
 We have previously indicated that there is a very strong 
 proof that ordinary matter possesses the property of emitting 
 characteristic radiations which are able to ionize a gas. Such 
 results suggest that there is an extremely slow transformation of 
 matter of a type similar to that shown by the radioactive bodies. 
 It is not necessary to suppose that the a particles from all types 
 of matter is the same mass. For example, hydrogen may be 
 expelled from some bodies instead of helium. The experimen- 
 tal observation that the a particle loses its power of acting on a 
 photographic plate and of ionizing the gas when its velocity 
 falls to about 8 x 10 8 cms. per second is of great importance in 
 this connection. There is no doubt that if a particles were ex- 
 pelled from matter below this velocity they would produce very 
 little if any electrical effect. It is certainly a matter of remark 
 
 1 Bumstead: Phil. Mag., Feb., 1906. 
 
276 RADIOACTIVE TRANSFORMATIONS 
 
 that the average a particle from radioactive bodies is projected 
 at a speed less than twice this minimum velocity. It appears 
 by no means improbable that the so-called radioactive bodies 
 may differ from ordinary matter mainly in their power of ex- 
 pelling a particles above this critical speed. Ordinary matter 
 which produces extremely weak ionizing effects might be emit- 
 ting a particles at a rate comparable with uranium, but yet, if 
 their power of expulsion was less than this critical value, it 
 would be difficult to detect their presence. 
 
 Such considerations show that it is by no means necessary 
 to suppose that the transformation of matter should always be 
 accompanied by the intense electrical and other effects exhibited 
 by the radioactive bodies proper. Matter may be undergoing 
 slow atomic transformation of a character similar to radium, 
 which would be difficult to detect by our present methods. 
 
INDEX. 
 
INDEX 
 
 a RATS 
 
 discovery of, 1 1 
 
 production of scintillations by, 12 
 
 nature of, 20 
 
 connection of a particle with helium, 
 
 181 and 184 
 
 properties of, 219 et seq. 
 velocity and mass of, 220 and 229 
 homogeneous character of, from radium 
 
 C, 220 
 
 magnetic deflection of, 223 et seq. 
 retardation of velocity in passing 
 
 through matter, 224 et seq. 
 connection between photographic action 
 
 and velocity, 225 
 minimum velocity to produce ionization, 
 
 225 
 
 phosphorescent effects produced by, 226 
 electrostatic deflection of, 227 
 deflection of, from thorium B, 229 
 scattering of, 230 
 magnetic deflection of, from a thick 
 
 layer of radium, 232 
 absorption of, by matter, 235 et seq. 
 law of absorption of, 236 et seq. 
 connection between ionization and ab- 
 sorption, 236 et seq. 
 ionization curves of radium and radium 
 
 C, 239 
 
 range of, from radium, 240 
 ionization curve of, for thick layer of 
 
 radium, 243 
 charge carried by, 245 
 heating effect of, 247 et seq. 
 action of, in producing hydrogen and 
 
 oxygen, 253 
 
 summary of properties of, 254 
 causes of expulsion of, 266 et seq. 
 physical need of effects of, 268 et seq. 
 action of, in producing luminosity in 
 
 gases, 274 
 Abraham 
 
 electrical mass, 10 
 
 Absorption 
 
 of a rays by matter, 235 et seq. 
 
 connection of, with ionization, 236 
 
 range in air of a rays, 239 et seq. 
 
 connection between, and atomic weight, 
 
 243 
 Actinium 
 
 discovery of, 9 and 164 
 
 rapid escape of emanation from, 1GJ> 
 
 radiations from, 166 
 
 analysis of active deposit of, 166 
 
 separation of actinium X from, 167 
 
 separation of radioactinium from, 107 
 
 transformation products of, 168 
 Actinium A 
 
 period and properties of, 166 
 Actinium B 
 
 period and properties of, 166 
 
 electrolysis of, 167 
 
 volatilization of, 167 
 Actinium X 
 
 separation of, 167 
 
 period and properties of, 167 
 Adams 
 
 radioactivity of spring water, 207 
 Age 
 
 of radium, 148 
 
 of earth's internal heat, 213 et seq. 
 
 of radioactive minerals, 187 et seq. 
 Allan, S. J. 
 
 radioactivity of snow, 200 
 Allan, S. J. and Rutherford 
 
 radioactivity of atmosphere, 199 
 Allen, H. S. and Lord Blythswood, 201 
 
 radium emanation in hot springs, 201 
 Atmosphere 
 
 radioactivity of, 196 et seq. 
 
 excited activity from, 198 et seq. 
 
 presence of emanations of radium and 
 thorium in, 199 
 
 diffusion of emanations into, 203 
 
 amount of radium emanation in, 204 et 
 seq. 
 
280 
 
 INDEX 
 
 penetrating radiation present in, 207 
 electrical state of, 208 et seq. 
 Atomic weight 
 of radium, 8 
 of products of radium, 1 93 
 
 /8 RATS 
 
 discovery of, 9 
 nature of, 10 
 
 variation of mass with velocity, 10 
 electroscope for measurement of, 29 
 emission of, by products, 169 
 connection of, with atomic explosion, 169 
 Barnes and Rutherford 
 
 heating effect of radium emanation, 90 
 heating effect of products of radium, 
 
 247 et seq. 
 connection of a rays with heating effect, 
 
 247 et seq. 
 Barrell 
 
 production of radium in interior of 
 
 earth, 194 
 Becquerel, H. 
 
 discovery of radioactivity of uranium, 5 
 mass and velocity of the particle, 10 
 separation of UrX, 162 
 magnetic deflection of a rays, 221 
 curvature of path of a rays in a mag- 
 netic field, 233 
 Blanc 
 
 presence of thorium emanation in 
 
 spring water, 201 
 Blythswood, Lord, and Allen, II. S. 
 
 emanation in hot springs, 201 
 Bodlander and Ruuge 
 
 evolution of gases from radium so- 
 lutions, 252 
 Boltwood 
 
 amount of radium in minerals, 152 et 
 
 seq. 
 
 production of radium by uranium, 158 
 lead as final product of radium, 191 
 products of transformation of radio- 
 elements, 192 
 Boltwood and Rutherford 
 
 amount of radium in minerals, 156 
 position of actinium in radioactive 
 
 series, 177 
 Bragg 
 
 scattering of a and )8 rays, 230 
 magnetic deflection of a rays, 233 
 absorption of a rays, 233 et seq. 
 
 Bragg and Kleeman 
 
 absorption of a rays, 236 et seq. 
 
 iouization curves of rays from radium 
 
 240 et seq. 
 Bronson 
 
 direct deflection electrometer, 33 
 
 effect of temperature on active deposit 
 
 of radium, 118 
 Brooks, Miss 
 
 volatility of radium B, 115 
 
 decay curves of excited activity of 
 
 actinium, 166 
 Brooks and Rutherford 
 
 diffusion of radium emanation, 82 
 Bumstead 
 
 presence of thorium emanation iu 
 atmosphere, 199 
 
 disintegration of atoms by X-rays, 275 
 Bumstead and Wheeler 
 
 period of radium emanation, 72 
 
 diffusion of radium emanation, 82 
 
 radium emanation in atmosphere, 199 
 Burton and McLennan 
 
 radium emanation in petroleum, 202 
 
 CAMPBELL 
 
 radioactivity of ordinary matter, 217 
 Changes. (See Transformations) 
 Charge 
 
 carried by the ions, 3 
 
 carried by the o particle, 184 et seq. 
 
 carried by the a rays, 245 
 Collie and Ramsay 
 
 Spectrum of emanation, 89 
 Collision 
 
 ionization of a particle by, 184 
 
 number of ions produced by, 271 
 Concentration 
 
 of excited activity on the negative 
 
 electrode, 45 
 Condensation 
 
 of water round the ions, 3 
 
 of radium and thorium emanations, 76 
 
 et seq. 
 Conductivity 
 
 of air in closed vessels, 197 
 Conservation of radioactivity 
 
 examples of, 64 
 Cooke, H. L. 
 
 penetrating rays from the earth, 207 
 Crookes, Sir W. 
 
 discovery of scintillations, 12 
 
INDEX 
 
 281 
 
 separation of UrX, 161 
 Crystallization 
 
 effect of, on activity of uranium, 163 
 Curie, Mme. 
 
 separation of radium and polonium, 7 
 
 period of polonium, 1 44 
 
 absorption of a rays from polonium, 236 
 Curie, P. 
 
 period of radium emanation, 72 
 Curie, P. and Mme. 
 
 separation of radium, 7 
 
 discovery of excited activity, 12 and 95 
 Curie and Danne 
 
 diffusion of radium emanation, 82 
 
 volatility of active deposit of radium, 
 
 117 
 Curie and Dewar 
 
 production of helium by radium, 182 et 
 
 seq. 
 Curie and Laborde 
 
 heating effect of radium, 13 and 247 
 Current 
 
 through gases, 24 et seq. 
 
 variation of, with voltage, 26 
 
 measurement of, by electroscope, 28 
 ct seq. 
 
 measurement of, by electrometer, 32 
 
 DADOURIAN 
 
 presence of thorium emanation in soil, 
 
 199 
 Danne 
 
 presence of radium without uranium, 
 
 155 
 Danne and Curie 
 
 diffusion of radium emanation, 82 
 decay curves of active deposit of radium, 
 
 111 
 effect of temperature on active deposit 
 
 of radium, 117 
 Debierne 
 
 separation of actinium, 9 
 
 production of helium from actinium, 
 
 183 
 Deposit, active, of actinium 
 
 analysis and properties of, 166 
 Deposit, active, of radium 
 rapid transformations of, 95 
 slow transformations of, 122 et seq. 
 Deposit, active, of thorium 
 
 connection of, with emanation, 45 et seq, 
 analysis of, 48 et seq. 
 
 Des Coudres 
 
 magnetic and electric deflection of a 
 
 rays, 221 
 Deslandres 
 
 spectrum of helium from radium, 182 
 Dewar and Curie 
 
 production of helium by radium, 182 
 Diffusion 
 
 of radium emanation, 81 
 
 of actinium emanation, 165 
 Disintegration 
 
 account of theory of, 
 
 list of products of, 169 
 
 helium a product of, 1 79 et seq. 
 
 of matter in general, 217 
 
 possible causes of, 266 et seq. 
 Dissipation of charge 
 
 apparatus for measurement of, 208 
 
 effect of conditions on rate of, 212 
 Dolezalek 
 
 electrometer, 31 
 Dorn 
 
 discovery of radium emanation, 70 
 
 EARTH 
 
 radioactivity of, 196 et seq. 
 
 internal heat of, 213 et seq. 
 
 amount of radium in earth, 215 
 Ebert 
 
 ionization apparatus, 209 
 
 number of ions per c.c. of atmosphere, 209 
 Ebert and Ewers 
 
 emanation from the earth, 200 
 Electrolysis 
 
 of active deposit of thorium, 53 
 
 of ThX, 65 
 
 of active deposit of actinium, 167 
 Electrometer 
 
 use of, 31 
 
 Dolezalek, 31 
 
 direct reading modification, 33 
 Electron 
 
 discovery of, 2 
 
 variation of mass of, with speed, 10 
 
 identity of $ particle with, 10 
 
 number of electrons emitted from oi:e 
 gram of radium, 22 
 
 emission of slow moving electrons from 
 emanation, 121 and 245 
 
 theory of matter, 258 et seq. 
 
 radiation from, 261 
 
 models of atoms, 263 
 
282 
 
 INDEX 
 
 Electroscopes 
 
 construction and use of, in radioactive 
 
 measurement, 28 
 Elster and Geitel 
 
 discovery of scintillations, 1 2 
 
 radioactivity of earth and atmosphere, 
 16 and 197 et seq. 
 
 emanation in earth and atmosphere, 200 
 
 radioactivity of soils, 202 <>t seq. 
 
 effect of meteorological conditions oil 
 
 radioactivity of atmosphere, 203 
 Emanation 
 
 discovery and properties of, 12 
 
 emission of, from elements, 22 
 Emanation of actinium 
 
 discovery arid properties of, 9 and 164 
 
 experimental illustration of, 165 
 Emanation of radium 
 
 discovery and properties of, 70 et seq. 
 
 occlusion of, 71 
 
 period of decay of, 71 
 
 rate of production of, 73 et seq. 
 
 condensation of, 76 et seq. 
 
 diffusion of, 81 et seq. 
 
 physical and chemical properties of, 84 
 
 volume of, 85 
 
 spectrum of, 89 
 
 heat emission of, 90 and 247 
 
 summary of properties of, 93 
 
 production of helium from, 179 et seq. 
 
 presence of, in earth and atmosphere, 
 196 et seq. 
 
 amount of, in atmosphere, 204 et seq. 
 
 heating of o rays from, ,247 et seq. 
 
 gases produced by, 253 
 Emanation of thorium 
 
 properties of, 38 et seq. 
 
 conditions of escape of, 39 
 
 chemical nature of, 40 
 
 rate of transformation of, 41 
 
 condensation of, 40, 80 
 
 connection of, with active deposit, 46 
 et seq. 
 
 production of, by ThX, 63 
 Emanmm 
 
 (See Actinium) 
 Eve 
 
 infection of laboratories by radium, 135 
 
 amount of radium emanation in atmos- 
 phere, 204 
 
 collecting distance of charged wire in 
 atmosphere, 207 
 
 ionization produced by the radium ema- 
 nation, 211 
 Ewers and Ebert 
 
 emanation from the soil, 200 
 Excited radioactivity 
 
 (See Active deposit) 
 
 7 RAYS 
 
 discovery of, 11 
 
 nature of, 20 
 
 connection of, witli (3 rays, 25 
 
 measurement of, 29 
 
 evolved by radium, 253 
 
 Gates 
 
 effect of temperature on active deposit 
 of thorium, 54, 117 
 
 Geitel 
 natural ionization of air, 196 
 
 Geitel and Elster 
 discovery of scintillations, 12 
 radioactivity of earth and atmosphere, 
 
 16 and 197 et seq. 
 
 emanation in earth and atmosphere, 200 
 radioactivity of soils, 202 et seq. 
 effect of meteorological conditions on 
 radioactivity of atmosphere, 203 
 
 Giesel 
 
 separation of radium, 9 
 separation of emanium, 9, 165 
 deflection of & rays, 9 
 emanation from actinium, 165 
 separation of actinium X, 167 
 phosphorescent light of emanium, 274 
 
 Gockel and Zolss 
 
 relation between potential gradient and 
 dissipation in atmosphere, 212 
 
 Godlewski 
 
 occlusion of radium emanation, 76 
 diffusion of uranium X, 163 
 radiations Irom actinium, 166 
 separation oi actinium X, 167 
 
 Goldstein 
 
 canal rays, 18 
 
 Gonders, Hofmann, and Wolfl 
 experiments with radiolead, 145 
 
 HAHN 
 
 separation of radiothorium, 68 
 separation of radioactinium, 168 
 magnetic deflection of a rays from 
 radiothorium, 229 
 
INDEX 
 
 283 
 
 Heat 
 
 rate of emission of, from radium, 13 
 rate of emission of, from emanation, 90 
 emitted by products of radium, 247 et 
 seq. 
 
 Heaviside 
 electrical mass, 10 
 
 Helium 
 
 history of discovery of, 179 
 production of, by radium, 181 
 connection of, with a particle, 183 
 rate of production of, by radium, 186 
 
 Hillebrande 
 
 gases liberated from radioactive miner- 
 als, 179 
 
 analysis of radioactive minerals, 190 
 et seq. 
 
 Himstedt 
 
 emanation from thermal springs, 201 
 
 Hofmann, Gonders, and Wolfl 
 experiments with radiolead, 145 
 
 Huggins, Sir W., and Lady 
 
 spectrum of phosphorescent light of 
 radium, 274 
 
 IONIZATION 
 
 production of, by X rays, 3 
 
 production of, by radioactive sub- 
 stances, 5 
 
 influence of theory of, on development 
 of radioactivity, 18 
 
 methods of measurement of, 27 et seq. 
 
 produced by a ray, 235 et seq. and 271 
 
 variation of, witli distance for o particle, 
 237 
 
 curves for radium, 239 et seq. 
 
 rate of production of, in various gases, 
 244 
 
 of water by radium rays, 253 
 Ions 
 
 condensation of water upon, 3 
 
 charge carried by, 3 
 
 recombination of, 25 
 
 delicacy of electrical method for de- 
 tection of, 35 
 
 number present per c.c. in atmosphere, 
 208 
 
 rate of production of, in air, 209 
 
 presence of slow moving in air, 210 
 
 rate of, production of by radium, 253 
 
 absorption of energy in production of, 
 271 
 
 KAUFMANN 
 
 variation of mass of electron with speed, 
 
 10 
 Kelvin 
 
 internal heat of earth, 213 
 
 models of atoms, 263 
 Kleeman and Bragg 
 
 absorption of a rays, 236 et seq. 
 
 ionization curves for the a rays from 
 
 radium, 240 et seq. 
 Kunzite. 
 
 phosphorescence of, 78 
 
 LABORDE and Curie 
 
 heat emission of radium, 13, 247 
 Langevin 
 
 presence of slow moving ions in atmos- 
 phere, 210 
 Larmor 
 
 electronic theory, 4 
 
 radiation from moving electron, 261 
 Lead, radioactive 
 
 (See Radiolead) 
 Lenard 
 
 cathode rays, 1 
 Lerch, von 
 
 electrolysis of active deposit of thorium, 53 
 
 electrolysis of ThX, 65 
 Life 
 
 of radium, 148 et seq. 
 
 of radioelements, 175 
 Lockyer, Sir Norman 
 
 discovery of helium in the sun, 179 
 Lorentz 
 
 electronic theory, 4 
 Luminosity 
 
 spectrum of phosphorescent light from 
 radium and emanium, 274 
 
 MACIIE and von Schweidler 
 
 velocity of ions in the atmosphere, 210 
 Mackenzie 
 
 magnetic and electric deflection of a 
 rays, 221 
 
 dispersion of a rays from radium in a 
 
 magnetic field, 232 
 Makower 
 
 diffusion of the radium emanation, 82 
 Marckwald 
 
 preparation of a radium amalgam, 8 
 
 separation of radiotellurium, 9, 140 
 
 period of radiotellurium, 139 
 
284 
 
 INDEX 
 
 Mass 
 
 of electron, 2, 10 
 
 apparent mass of a particle, 184 
 Materials 
 
 radioactivity of ordinary, 217 
 McClelland 
 
 secondary radiation due to B and y rays, 
 
 246 
 McClung 
 
 range of ionization of rays from radium 
 
 C, 239 
 McCoy 
 
 activity of uranium minerals, 152 
 McLennan 
 
 radioactivity of snow, 200 
 
 penetrating radiation from earth, 217 
 McLennan and Burton 
 
 emanation in petroleum, 202 
 Methods of measurement 
 
 in radioactivity, 23 et seq. 
 
 comparison of photographic and electri- 
 cal, 24 
 
 description of electrical, 24 et seq. 
 Meyer and Schweidler 
 
 period of radiotellurium, 139 
 
 radioactive transformations of radio- 
 lead, 146 
 
 effect of crystallization of uranium on 
 
 its activity, 163 
 Minerals, radioactive 
 
 final products present in, 192 
 
 age of, 187 
 
 amount of lead and helium present in, 
 
 191 
 Moore and Schlundt 
 
 separation of ThX, 65 
 
 OCCLUSION 
 
 of emanation in radium and thorium 
 
 compounds, 74 
 Oxygen 
 
 production of, in radium solutions, 253 
 
 PEGRAM 
 
 electrolysis of active products from 
 thorium, 53 
 
 heating effect of thorium, 252 
 Phosphorescence 
 
 produced by radium, 23 
 
 produced by radium emanation in sub- 
 stances, 78 
 
 connection of, with ionizatiou, 226 
 
 spectrum of phosphorescent light of 
 
 radium compounds, 274 
 Photographic 
 
 method of measurement, 24 
 
 connection of photographic action with 
 
 ionization, 226 
 Pohl and Walter 
 
 luminosity of gases exposed to a rays> 
 
 274 
 Polonium 
 
 separation of, 7 
 
 period of, 143 
 
 identity of, with radiotellurium and 
 radium F, 143 
 
 slow moving electrons from, 245 
 Products, radioactive 
 
 from thorium, 67 
 
 from radium, 130 
 
 from uranium, 163 
 
 from actinium, 168 
 
 amount of, in radium, 142 
 
 properties of, 173 
 
 chemistry of, 173 
 
 RADIATIONS 
 
 from active bodies, 19 et seq. 
 
 from different active products, 1 70 et 
 seq. 
 
 connection of, with heat emission, 247 
 Radioactininm 
 
 separation and period of, 168 
 Radiolead 
 
 discovery of, 9 
 
 radioactive analysis of, 144 
 
 connection of, with radium D, 145 
 Radiotellurium 
 
 discovery of, 9 
 
 amount of, in radium minerals, 141 
 
 period of, 139 
 
 identity with radium F and polonium,. 
 
 139 
 Radiothorium 
 
 discovery and properties of, 68 
 Radium 
 
 discovery of, 7 et seq. 
 
 properties of, 8 et seq. 
 
 amount of, in uranium minerals, 8 
 
 radiations from, 9 and 20 
 
 emanation from, 12 
 
 properties of emanation from, 70 et seq* 
 
 recovery of activity of, 73 
 
 occlusion of emanation in, 74 
 
INDEX 
 
 285 
 
 condensation of emanation of, 76 et seq. 
 
 diffusion of emanation of, 81 et s^q. 
 
 volume of emanation of, 85 et seq. 
 
 spectrum of emanation of, 89 et seq. 
 
 heat emission of emanation of, 90 et seq. 
 
 excited activity produced by, 95 et seq. 
 
 decay curves of excited activity of, 99 
 et seq. 
 
 theory of successive changes in, 104 et 
 seq. 
 
 analysis of active deposit of, 104 et srq. 
 
 effect of temperature on active deposit 
 of, 116 et seq. 
 
 slow transformation products of, 122 et 
 seq. 
 
 properties of radium D, E, and F, 122 
 et seq. 
 
 variation of activity of, over long inter- 
 val, 137 et seq. 
 
 identity of radium F with radiotellurium, 
 138 et seq. 
 
 identity of radium F with polonium, 143 
 et seq. 
 
 connection of, with radiolead, 144 
 
 life of, 148 et seq. 
 
 origin of, 151 et seq. 
 
 amount of, in minerals, 155 et seq. 
 
 growth of, in uranium, 157 et seq. 
 
 connection of, with uranium and actin- 
 ium, 176 et seq. 
 
 production of helium from, 179 et seq. 
 
 age of radium minerals, 187 et seq. 
 
 chemical constitution of, 193 et seq. 
 
 presence of, in earth and atmosphere, 
 196 et seq. 
 
 amount of, in atmosphere, 204 et seq. 
 
 heating of earth by, 213 et seq. 
 
 properties of a rays from, 219 et seq. 
 
 ionization produced by a rays from, 235 
 et seq. 
 
 heating of a rays from, 247 et seq. 
 
 gases evolved by, 253 et seq. 
 
 processes occurring in, 256 et seq. 
 Radium A 
 
 nomenclature, 95 
 
 analysis of, 101 et seq 
 
 effect of presence of, in decay curves, 
 112 et seq. 
 
 connection of, with radium B, 1 14 et seq. 
 Radium B 
 
 analysis of, 101 et seq. 
 
 effect of temperature on, 1 1 6 et seq. 
 
 true period of, 119 
 
 rays emitted from (footnote, 115) 
 Radium C 
 
 analysis of, 101 et seq. 
 
 emission of j8 and y rays by, 102 et seq. 
 
 effect of temperature on, 116 et seq. 
 
 true period of, 119 
 
 violent disintegration of, 169 et seq. 
 
 properties of a rays from, 222 et seq. 
 Radium D 
 
 nomenclature, 122 
 
 analysis of, 122 et seq. 
 
 effect of temperature on, 126 
 
 properties of, 129 
 
 period of, 131 et seq. 
 
 presence of, in old radium, 136 
 
 connection of, with radiolead, 144 et seq. 
 Radium E 
 
 analysis of, 122 et seq. 
 
 period of, 125 
 
 effect of temperature on, 126 
 
 connection of, with radium F, 128 
 Radium F 
 
 analysis of, 122 et seq. 
 
 rise of activity due to, 124 
 
 effect of temperature on, 126 
 
 separation of, by bismuth, 127 
 
 period of, 127 
 
 variation of activity due to, 133 
 
 presence of, in radium, 136 
 
 identity of, with radiotellurium, 138 et 
 seq. 
 
 identity of, with polonium, 142 et.seq. 
 
 amount of, in uranium, 140 et seq. 
 Rain 
 
 radioactivity of, 200 
 Ramsay, Sir W. 
 
 discovery of helium, 179 
 
 atomic weight of helium, 180 
 
 amount of helium in atmosphere, 180 
 Ramsay and Collie 
 
 spectrum of radium emanation, 89 
 Ramsay and Soddy 
 
 volume of emanation from radium, 87 
 
 production of helium by radium, 181 
 
 rate of production of helium by radium, 
 186 
 
 evolution of gases from radium, 253 
 Rayless transformation 
 
 character of, 171 
 Rontgen 
 
 discovery of X-rays, 1 
 
286 
 
 INDEX 
 
 Runge and Bodlander 
 
 evolution of gases from radium, 253 
 
 SACK OR 
 
 period of radium emanation, 72 
 Schlundt and R. B. Moore 
 
 separation of ThX, 65 
 Schmidt 
 
 discovery of radioactivity of thorium, 7 
 Schmidt, W. C. 
 
 emission of ft rays from radium B, 115 
 Schuster 
 
 number of ions per c.c. of atmosphere, 
 
 209 
 
 Schweidler and Meyer 
 period of radiotellurium, 139 
 radioactive analysis of radiolead, 146 
 effect of crystallization of uranium on 
 
 its activity, 1 63 
 Schweidler and von Mache 
 
 velocity of ions in the atmosphere, 210 
 Scintillations 
 
 discovery of, in zinc sulphide screen, 
 
 12 
 use of, to determine range of a particles, 
 
 226 
 
 connection of, with ionization, 226 
 Searle 
 
 electrical mass, 10 
 Secondary rays 
 from radioactive bodies, 20 
 emission of, in form of slow moving 
 
 electrons, 245 
 Simpson 
 
 effect of meteorological conditions on 
 amount of emanation in atmosphere, 
 212 
 
 Slater, Miss 
 effect of temperature on active deposit 
 
 of thorium, 53 
 emission of slow moving electrons from 
 
 emanations, 121 
 Snow 
 
 radioactivity of, 200 
 Soddy 
 
 production of radium by uranium, 158 
 Soddy and Ramsay 
 
 volume of emanation, 87 
 production of helium by radium, 181 
 rate of production of helium by radium, 
 
 186 
 evolution of gases by radium, 253 
 
 Soddy and Rutherford 
 development of disintegration theory, 
 
 13 et seq. 
 emanating power of thorium com 
 
 pounds, 39 
 chemical nature of thorium emanation, 
 
 40 
 
 separation of ThX, 56 
 period of radium emanation, 71 
 condensation of emanations, 76 et seq, 
 physical and chemical properties of 
 
 emanations, 84 
 
 origin of helium in minerals, 181 
 Spectrum 
 
 of radium emanation, 89 
 
 of phosphorescent light of radium 
 
 bromide, 274 
 Strutt 
 
 amount of radium in minerals, 152 
 radium emanation in spring water, 201 
 connection of thorium and uranium, 177 
 radioactivity of ordinary matter, 211 
 
 TEMPERATURE 
 effect of, on active deposit of thorium, 
 
 54 
 effect of, on active deposit of radium, 
 
 116 and 126 
 effect of, on active deposit of actinium, 
 
 166 
 Thomson, J. J. 
 
 cathode rays and electrons, 2 
 
 charge carried by an ion, 3 
 
 electrical mass, 10 
 
 emission of slow moving electrons with 
 
 a particles, 120 
 
 radioactivity of well water, 201 
 charge carried by the a rays, 245 
 electronic model of atom, 264 
 radiation of energy from the atom, 265 
 
 and 267 
 Thorium 
 
 discovery of radioactivity of, 6 
 radiations from, 32 
 emanation from, 38 et seq. 
 emanating power of compounds of, 39 
 excited radioactivity from, 45 et seq. 
 transformation products of, 48 et seq. 
 separation of ThX from, 56 et seq. 
 separation of radiothorium from, 68 
 possible connection of, with uranium., 
 176 et seq. 
 
INDEX 
 
 287 
 
 Thorium A 
 
 period ami properties of, 48 et seq. 
 
 electrolysis of, 53 
 
 effect of temperature on, 54 
 Thorium B 
 
 period and properties of, 48 et seq. 
 
 electrolysis of, 53 
 
 effect of temperature on, 54 
 
 complexity of, 242 
 Thorium X 
 
 discovery of, 13, 56 et seq. 
 
 separation of, 56 
 
 decay and recovery curves of, 57 et seq. 
 
 emanation from, 62 
 
 irregularities in decay of, 63 
 
 methods of separation of, 65 
 
 table of changes of, 67 
 Townsend 
 
 charge on an ion, 3 
 Transformations 
 
 general theory of, 14 
 
 connection of law of decay of activity 
 with, 42 
 
 of thorium, 46 et seq. 
 
 mathematical theory of, 50 et seq. 
 
 of radium, 94 et seq, 122 et seq. 
 
 of uranium, 161 
 
 of actinium, 164 et seq. 
 
 connection between, of radioelements, 
 168, 176 
 
 ray less, 171 
 
 properties of products of, 173 
 
 position of helium as product of, 179 et 
 seq. 
 
 causes of atomic, 267 
 
 processes of, 269 
 
 of inactive matter, 275 
 
 URANIUM 
 
 discovery of radioactivity of, 5 
 amount of radium in uranium minerals, 
 
 8 and 155 
 
 connection of, with radium, 151 
 growth of radium in, 157 
 
 transformation products of, 161 
 
 separation of UrX 161 
 
 effect of crystallization on radioactivity 
 
 of, 163 
 Uranium X 
 
 separation of, 161 
 radiations from, 161 
 
 effect of crystallization on distribution 
 of, 161 
 
 VELOCITY 
 
 of a particle from radium C, 229 
 retardation of, in passing through 
 
 matter, 224 
 connection of, with range of ionization, 
 
 225 
 Villard 
 
 discovery of 7 rays, 1 1 
 
 WALTER and Pohl 
 
 luminosity of gases exposed to o rays, 274 
 Weichert 
 
 nature of X-rays, 2 
 Wheeler and Burnstead 
 
 period of radium emanation, 72 
 
 diffusion of radium emanation, 82 
 
 radium emanation in atmosphere, 199 
 Whetham 
 
 production of radium from uranium, 158 
 Wien 
 
 canal rays, 18 
 Wilson, C.T. R. 
 
 ionic nuclei, 3 
 
 electroscope, 29 
 
 natural iouization of air, 197 
 
 radioactivity of rain and snow, 200 
 Wolfl, Hofmann, and Gonders 
 
 experiments in radiolead, 145 
 Wood 
 
 radioactivity of ordinary matter, 217 
 
 ZOLSS and Gockel 
 
 relation between potential gradient and 
 rate of dissipation in atmosphere, 212 
 
 OF THE 
 
 UNIVERSITY I) 
 
 ~ OF 
 
 &PF 
 
DATE 
 
 14 DAY USE 
 
 RETURN TO DESK FROM WHICH BORROWED 
 
 LOAN DEPT. 
 
 This book is due on the last date stamped below, or 
 
 on the date to which renewed. 
 Renewed books are subject to immediate recall. 
 
 1-0 
 
 MAR 
 
 - - 
 
 LOAN 
 
 MAY 71968 
 
 1969 OU 
 
 DEC 2 1197 
 
 LD 21A-50m-3,'62 
 (C7097slO)476B 
 
 General Library 
 
 University of California 
 
 Berkeley 
 
UNIVERSITY OF CALIFORNIA LIBRARY