You are on page 1of 167



J. J.


D.Sc., LL.D., PH.D., F.R.S.






Published, March, 1904


In the year 1883 a legacy of eighty thousand dollars left to the President and Fellows of Yale College in the city of New Haven, to be held in trust, as a of their beloved and gift from her children, in memory honored mother Mrs. Hepsa Ely Silliinan. On this foundation Yale College was requested and directed to establish an annual course of lectures designed to illustrate the presence and providence, the wisdom and goodness of God, as manifested in the These were to be designated natural and moral world. as the Mrs. Hepsa Ely Silliinan 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 foundation, and that the subjects should be selected rather from the domains of natural science


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. Sillimau. The memorial fund came into the possession of the Corporation of Yale University in the year 1902; and the present volume constitutes the first of the series of memorial


it seemed to me that a consideration of the bearthis relationship ing of recent work on might be suitable. such as the study and discovery and Rontgen Rays and Radio-active Substances. Aug. Cambridge. J. 1903.PREFACE In these Lectures given at Yale University in May. In choosing a subject for the Silliman Lectures. that the solution of the one would supply that of the other. especially as such a discussion suggests multitudes of questions which would furnish admirable subjects for further investigation by some of my hearers. J.. I have attempted to discuss the bear- ing of the recent advances made in Electrical Science on our views of the Constitution of Matter and the Nature of Electricity. ÆTHE RF ORCE . THOMSON. has been the very especial degree in which they have involved the relation between of Cathode Matter and Electricity. 1903. two questions which are probably so intimately connected. A characteristic feature of recent Electri- cal Researches.


is as old almost as the science itself . The two-fluid theory explains the phenomena of electro-statics by supposing that in the universe there are two fluids. and of the connection between electrical and ordinary matter which have been suggested by the investigations. results of recent The progress of electrical science has been greatly promoted of electricity. of the MY object in these lectures processes going on in the electric field. I mean the theories known as the two.ELECTRICITY AND MATTER CHAPTER I REPRESENTATION OF THE ELECTRIC FIELD BY LINES OF FORCE is to put before you and in as simple untechnical a manner as I can some views as to the nature of electricity. uncreatable and indestruc- ÆTHE RF ORCE . by speculations as to the nature it hardly possible to overestimate the services rendered by two theories Indeed.and the one-fluid theories of electricity.

as do of the negative fluid. contains more positive fluid than negative it is positively electrified. so as to afford a basis for the explanation of gravitation. the other and electrical phenomena are explained by folascribing to the fluids the The particles of the positive lowing properties. The bile supposed to be exceedingly moand able to pass with great ease through confluids are ductors. the particles of the positive fluid tract those of the negative fluid. m to the repulsion between two m and m of the same sign. called positive. contains equal quantities Since the fluids are uncreatable if it ÆTHE RF ORCE . placed in the same position as the previous charges. one of these fluids negative electricity. with forces varying inversely other fluid repel each as the square of the distance also the particles between them. on the at- other hand. are in one form of the theory supposed to be exactly equal charges. The state of electrification of a body is determined by the difference between the quantities of the two electric fluids contained by it if it . it is uncharged. ELECTRICITY AND MATTER whose presence gives is rise to electrical effects. of opposite signs between two charges. In anis other development of the theory the attraction supposed to slightly exceed the repulsion. The attraction and m.2 tible.

say zinc and copper. negative electricity. the appearance of the positive in one place must be accompanied by the departure of the same quantity from some other place. so ÆTHE RF ORCE . the zinc becoming positively.LINES OF FORCE 3 and fluid indestructible. with each other. On sist this view. fluids. so that the production of electrification of one sign must always be accompanied by the production of an equal amount of electrification of the opposite sign. the attraction of zinc for positive electricity being greater than that of copper. positive elec- tricity. are put electrified. He did this when two in contact what is known as contact electricity.e. but in the earlier form of the theory no action was contemplated between ordinary matter and the electric fluids it was not until a . the copper negatively Helmholtz sup- posed that there are forces between ordinary matter and the electric fluids varying for different kinds of matter. comparatively recent date that Helmholtz introduced the idea of a specific attraction between ordinary matter and the electric to explain i. The two supposed to exert forces on themselves and on each other. the electrical separation produced metals.. every : body is supposed to conlatter are of three things ordinary matter.

the is the part of the other taken by ordi- nary matter. we have to suppose that the mixture of the two fluids in equal proportions is something so devoid of physical properties that its existence has never been detected. of the two fluids exactly neutraliz- we regard these fluids as being than the mathematical anything more substantial us into difficulties if leads this and symbols + . it implies that if are equal quantities of the two added to the body. It gives no information about the amount of either. the particles of which are supposed to repel each other and attract the positive fluid. There is an indefiniteness about the two-fluid be illustrated by the consideratheory which may unelectriEed an of tion body. we regard them as physical fluids.4 that zinc ELECTRICITY AND MATTER when robs these metals are put in contact the its the of copper of some positive electricity. only one electric fluid. the equal quantities ing each other. not open to this objection. ÆTHE RF ORCE . The other fluid theory is is the one-fluid theory of Benjamin Franklin On this view there positive . for example. indeed. All that the twofluid theory tells us about such a body is that it contains equal quantities of the two fluids. If body will be unaltered.

intended merely to give .LINES OF FORCE just as the particles of the negative fluid 5 do on the two-fluid theory. is Matter when unelectrified supposed to be associated with just so much of the electric fluid that the attraction of the matter on a portion of the the same fluid electric fluid outside it is just sufficient to counteract the repulsion exerted on by is the electric fluid associated with the matter. both theories must give the same results and there can be nothing to ÆTHE RF ORCE . The services which the fluid theories have ren- dered to electricity are independent of the notion of a fluid with any physical properties the fluids were mathematical fictions. As which only involve the law of forces between electrified bodies. and served as the means by which the splendid mathematical development of the theory of forces varying inversely as the square of the distance which was inspired by the discovery of gravitation could be brought to bear on long as electrical we confine ourself to questions phenomena. a local habitation to the attractions and repulsions existing between electrified bodies. ter in a fluid is at On this view. and the simultaneous production of equal quantities of -{- and electricity. if the quantity of mat- body known the quantity of electric once determined.

while we have no reason to anticipate so great a difference on the two-fluid theory. maticians The physicists and mathe- who did most to develop the "Fluid Theories kind. shall. expect on Franklin's one-fluid theory. if that theory were modified by making the electric fluid correspond to negative instead of positive electricity. Let us take a case which has actually We have been able to measure the masses with given charges of electricity in gases at low pressures. and it has been found that the mass associated with a positive charge is imassociated mensely greater than that associated with a negaThis difference is what we should tive one. " confined themselves to questions of this and refined and idealized the conception of these fluids until any reference to their physical properties was considered almost indelicate. am by the results of the most recent researches with in the very infancy those enunciated of the subject.g ELECTRICITY AND MATTER decide between them. It is not until we investigate phenomena which involve the physical properties of the fluid that we can hope to distinguish between the rival fluid theories. arisen. I We be struck by the similarity between some of the views which we are led to take sure. by Franklin ÆTHE RF ORCE .

or if you prefer it. and have devoted much thought and labor to replacing it by some- thing involving mechanical continuity. Faraday was deeply influenced that matter cannot act by the axiom. dogma where it is not. almost unrivalled mathe- matical insight. who possessed. the bodies exerting the ÆTHE RF ORCE . This idea. Faraday. it and replace by one which would bring into between prominence some continuous connection was able to He forces. imply the idea of action at a distance. had had no training in analysis. so that the convenience of the idea of action at a distance for purposes of calculation had no chance of mitigating the repugnance he felt to the idea of forces acting far away from their base and with no physical connection with their origin. from their very nature. Pre-eminent among them is Faraday. I believe. is one which many of the greatest physicists have felt utterly unable to accept. He therefore cast about for some way of picturing to himself the actions in the electric field which would get rid of the idea of action at a distance. although has its convenience for mathematical analysis acceptable to made it many mathematicians.LINES OF FORCE 7 Faraday's Line of Force Theoi^y The fluid theories.

FIG. i. 1 . and as I believe its some time to the discussion and development of this conception of the electric field. to consideration of the lines of force Faraday by the round a bar on a smooth If iron filings are scattered surface near a in Fig. magnet they arrange themselves as well-marked lines can be traced run- ÆTHE RF ORCE . I shall use of this method. The method was suggested magnet. As I have continually to make powers and possibilities have shall devote been never adequately realized.g ELECTRICITY AND MATTER this do by the conception of lines of force.



ning from one pole of the magnet to the other

the direction of these lines at any point coincides with the direction of the magnetic force, while the
intensity of the force
tration of the lines.


by the concen-

Starting from any point in the field and travelling always in the direction of the magnetic force, we shall trace out a line which
will not stop until


reach the negative pole of

the magnet



such lines are drawn at


in the field, the space

through which the magnetic
with a system of

extends will be


giving the space a fibrous structure like that possessed by a stack of hay or straw, the grain of the
structure being along the lines of force.


spoken so far only of lines of magnetic force


same considerations

will apply to the electric field,

and we may regard the electric field as full of lines of electric force, which start from positively and end on negatively electrified bodies. Up to
this point the process has

been entirely geometri-

and could have been employed by those who looked at the question from the point of view of
action at a distance; to Faraday, however, the
lines of force

were far more than mathematical

they were physical realities. Faramaterialized the lines of force and endowed day




so as to explain the

them with physical properties


of the electric


Thus he


in a state of tension, and posed that they were Instead of an inthat repelled each other.
at a distance between tangible action
fied bodies,



Faraday regarded the whole space between the bodies as full of stretched mutually
repellent springs.

The charges

of electricity to

which alone an interpretation had been given on the fluid theories of electricity were on this view
just the ends of these springs,

and an


of fluid confined charge, instead of being a portion to the electrified body,

was an extensive arsenal

of springs spreading out in all directions to all parts of the

To make our
of view.

ideas clear on this point let us

consider some simple cases from Faraday's point

Let us


take the case of two bodies

with equal and opposite charges, whose lines of You notice that the force are shown in Fig. 2.
lines of force are

most dense along

A B, the


joining the bodies, and that there are more lines of force on the side of than on nearest to



the opposite side.
lines of force

Consider the effect of the



the lines are in a state of


and are pulling away at -4; as there




more pulling at than on the opposite


on the side nearest to
the pulls on



A toward

overpower those pulling away from B, so will tend to move toward B] it was in that




Faraday pictured to himself the attraction between oppositely electrified bodies.



Let us


consider the condition of one of the

curved lines of

such as


it is

in a state



of tension and will therefore tend to straighten

prevented from doing this and maintained in equilibrium in a curved position?
is it



can see the reason for this


we remember

that the lines of force repel each other and that

the lines are more concentrated in the region beand tween than on the other side of




thus the repulsion of the lines inside will be greater than the repulsion of those out-


and the


PQ will be bent outwards.


Let us suppose since the A B are positively start lines force from end on negatively electrified bodies. 3. spread out. the lines of force for which are shown in Fig. if A B present. ÆTHE RF ORCE . Consider now the effect of making the system of lines of force attached to A and B approach each other . uniformly in all directions. negative charges corresponding to the positive and B] let us suppose that these ones on A charges are a considerable distance away. 3.12 ELECTRICITY AND MATTER Let us now pass from the case of two oppositely electrified bodies to that of two and of similarly elec- trified ones. electrified. in the part of the field under consideration. so that the lines of force from were not would. positively and will travel away and the lines starting from A B to join some body or bodies possessing the FIG.

the lines of force between the plates are straight except near the edges of the plates . thus the pulls exerted on A in the rear by the lines of force will be greater than those in the front and the result will be that A will be pulled away from B. and mechanism producsame type as please regard the that which produced the attraction in the pre- we may if we repulsion between tions A and B as due to the attrac- on A and B of the complementary negative charges which must exist in other parts of the field. you will notice that PIG. that of two oppositely electrified plates. so that the lines of force will now be densest on the far side of A . this is what we should expect as the downward pressure exerted ÆTHE RF ORCE . 4.LINES OF FORCE since these lines repel each other the lines of force on the side of A nearest B will be pushed to the opposite side of A. The sion results of the repul- of the lines of force are clearly shown in the case represented in Fig. We notice that the ing this repulsion is of exactly the vious case. 4.

upward pressure For a line of force near the edge of the plate. can do this by introducing the If through the boundary of any small closed curve in the electric field we draw the lines of force. been descriptive rather than metrical it is. and the line of force will bulge out until its curvature and tension counter. So far our use of the lines of force has . and if we follow the lines back to the positively electrified surface from which they start and forward on to the negatively electrified surface on which they end. however. we can prove that the positive charge enclosed is by the tube at its origin equal to the negative charge enclosed by it at its end. method so as to make it We idea of tubes of force. these lines will form a tubular surface. the pressure of the lines of force below will exceed the pressure from those above. Let us call ÆTHE RF ORCE . however. By properly choosing the area of the small curve through which force. 4. we draw the lines of we may is the tube arrange that the charge enclosed by equal to the unit charge.J4 ELECTRICITY lines of force AND MATTER line in this part of by the exerted above a the field will be equal to the by those below it. act the squeeze from inside this bulging is very plainly shown in Fig. easy to develop the metrical.

as the termination of a We regard di- these Faraday tubes as having direction. the . number being reckoned algebraically i. we draw any closed surface then the difference between the number of Faraday tubes which pass out of the surface and those which pass in will be equal to the algebraic sum of the charges surface.. while those which pass through in the opposite direction are taken as negative.e. their rection being the same as that of the electric force.LINES OF FORCE 15 such a tube a Faraday tube positive electricity in the field as the origin then each unit of may be regarded electricity and each unit of negative Faraday tube. inside the surface. the lines which pass and the number and the passing through the area is the difference between positively the number passing through number passing through negatively. I have found the conception ÆTHE RF ORCE . is so that the positive direction from the positive If to the negative end of the tube. through in one direction being taken as positive. this sum is what Maxwell called the electric displace- ment through the What Maxwell called the electric displacement in any direction at a point is the number of Faraday tubes which pass through a unit area through the point drawn at right angles to that direction. For my own part.

we see that they are equivalent a tension \ NR along the direction of the electric force and an equal pressure in force. in addition to this tension.e. If drawn angles to we consider the effect of these tensions and pressure on a unit volume of the medium in the electric to field. all directions at right angles to that ÆTHE RF ORCE .. N being the density the at a hydrostatic of the Faraday tubes. Maxwell took up the question of the tensions and pressures in the lines of force in the electric field. i. By calculating the amount of these tensions he showed that the mechanical effects in the electrostatic field could be explained by supposing that each Faraday tube force exerted a tension equal to R. there was pass in the medium through which the tubes pressure equal to \NR. and carried the problem one step further than Faraday. and have latter many years abandoned the method. and that.IQ ELECTRICITY AND MATTER of to much more readily Faraday tubes to lend itself the formation of a mental picture of the procon in the electric field esses going electric than that of for displacement. being H the intensity of the electric force. number passing right through a unit area the electric force.

let us now proceed to the study of the effects produced by the motion of those tubes. contract to molecular dimensions and the repulsion they previously exerted on neighboring disappear. and suppose that being charged the plates are connected by a conducting wire. one with the positive other with negative after electricity. disappeared so tension. this tubes will therefore Consider the effect of F FIG. however. . when in a conductor. 5. This wire will pass through some of the outlying tubes these tubes. EFG. Let us begin with the consideration of a very simple case plates.LINES OF FORCE 17 Hitherto to be at Moving Faraday Tubes we have supposed the Faraday tubes rest. charged. Thus. and we shall ÆTHE RF ORCE . A that of two parallel and B. on a tube . but will EFG. own EF that PQ will no longer be in equilibrium. The repulsions due to those neighboring cut by G have now. PQ between the plates its PQ was originally in equilibrium under and the repulsion exerted by the tubes. more and more be pushed towards tubes will be pushed into E FG.

Thus. accompanied is. charged plate the plates. tween the plates are moving at right angles to themselves. the magequal to kiro-v. while the EF the tubes bedischarge of the plates is going on. Here we have two phenomena which do not take place in the steady electrostatic field. I have followed up the consequences of this supposition and have shown that.jg ELECTRICITY AND MATTER tween the plates toward have a movement of the whole set of tubes beG. this magnetic force is at right angles to the plane of 47r the paper and equal to times the intensity of if a. the other the existence of a magnetic force. this suggests that there is a connection between the two. this What physical ? effect accompanies result of con- movement of the tubes The is necting the plates rent of electricity by EF G as to produce a cur- flowing from the positively to the negatively charged plate through EF G we know. and that motion of the Faraday tubes is accompanied by the production of magnetic force. This force between a magnetic by . if the connection between the ÆTHE RF ORCE . or. the density of the charge of electricity on the plates and v the velocity with netic force is which the charge the current in the plate. one the movement of the Faraday tubes.

that variation in the electric displacement in a dielectric produces magnetic force. Faraday tubes must move up to or away from the place. and also to its direction of motion 6 is the angle between the Faraday tube and the direction in which it is moving. produces at tude is 4?r v sin 0. this view will account for Ampere's laws connecting current and magnetic force. the electric displacement at any place changes. ÆTHE RF ORCE . no such force produced by the gliding of a tube along its length. We see that it is only the motion of a tube at right angles to itself is which produces magnetic force.LINES OF FORCE 19 magnetic force and the moving tubes is that given below. and motion of Faraday tubes. follows at once from For. esis. The law connecting magnetic force with the motion of the Faraday tubes is as follows Faraday tube moving with velocity v at a point : A a magnetic force whose magniP. the direction of the magnetic force being at right angles to the P Faraday tube. by hypoth- implies magnetic force. . and for Faraday's law of the induction of currents. measured by the density of the Faraday tubes. Maxwell's great contribution to electrical theory. since the electric displacement is if this view.

They will be carried along with the sphere. OP and the direction of motion of according to the above rule. the is density of the Faraday tubes at so that if P TTp*' v is the velocity of the sphere. its If e is the charge on the sphere. If the velocity of the sphere is small compared with that of light then the Fara- day tubes uniformly will. the lines of magnetic force will thus ÆTHE RF ORCE . the direc- tion of the force will be at right angles to OP. 6 the angle between the sphere. centre. be distributed and in direction. as when the sphere radial is at rest. and at right to the direction of motion of angles the sphere. the magnetic force at P will be 6V * .20 ELECTRICITY AND MATTER Motion of a Charged Sphere We and apply these results to a very simple important case the steady motion of a shall charged sphere. then.

. 6 2 . energy for parts of the field outside the sphere. outside the -z sphere.- . Typr- Taking the sum of .instead of m. we know that in a unit volume of the field at a place where the magnetic force is H there are ~ O7T units of energy. the kinetic energy in the sphere is m v9 in addition to that we have the energy . sphere and their planes at right Thus. so that the whole kinetic energy of the is m (2u -\- -o- \ ~~ i 2 0*. a ÆTHE RF ORCE . a moving charge of will be accompanied by a magnetic The existence of a magnetic field implies energy. where a the radius of the sphere. m the mass of the sphere. or the energy the same as if the mass of the sphere were Thus. where /x is the magnetic permeability of the medium. electricity field. is we find that it amounts to If ~- 3a is . is a^v* i= all sin 2 . In the case of the moving sphere the energy per unit volume this at D P . in consequence of m -f- o 2u. having their centres on the path of the centre of the angles to this path. which as we have seen is od .LINES OF FORCE 21 be circles.

the amount by ÆTHE RF ORCE . volume of the sphere. its mass is increased by the mass of an equal volume of the liquid. ume of the liquid. as was shown by half the Green in 1833. since it re- shows that part of the mass of a I shall later charged sphere is due to its charge. Before passing on to this point. that the sphere behaves as if its mass were increased by that of a certain volitself. is This volume. which considerations on have to bring before you show that it is not impossible that the whole mass body may arise in the way. The first of these is the case of a sphere frictionless liquid. elongated body In the case of an like a cylinder. however. but also the liquid around it the consequence of this is. I should like to illustrate the increase which takes of a place in the mass of the sphere by some analogies drawn from other branches of physics. When moving through a the sphere moves it sets the fluid around tioned to its it moving with a velocity propor- own.22 ELECTRICITY AND MATTKK is the electric charge. so that to move the sphere we we have sphere not merely to move the substance of the . the mass of the sphere 9. its In the case of a cylinder moving at right angles to length.1 $ ~&^L^AAoL_ measured by -^. This is a very important sult.

The mass of such direction in which it is body depends on the moving. thus. but m -f- "^ The additional mo- mentum - v is not in the sphere. however. if it moving with the / velocity v. because the recognition of it makes the behavior of the electric field ÆTHE RF ORCE . but in the space surrounding the sphere. We have seen that in consequence of its is charge its mass is increased by -(* oa . being much smaller than a when the body moves point foremost when moving sideways. I want to bring the existence of this momentum before you as vividly and forcibly as I can. return to the moving electrified sphere. There is in this space ordinary mechanical momentum.LINES OF FORCE 23 which the mass rection in is increased depends is upon the di- which the body moving. to bear in mind that this momentum important not in any way different from ordinary mechanical momentum and can be given up to or taken from the momentum of moving bodies. whose resultant is v and whose direction is parallel to the diIt is is rection of motion of the sphere. Let us. momentum is not mv. the j v.

Now. brought into accordance with this law nize the existence of the field .e. but none in the body. consider ordinary matter. in the case of many electrical systems there .24 ELECTRICITY AND MATTER of a mechanical sys- to that entirely analogous take an example. The phenomenon however. for the only this momentum recognized on restricted is. so that direction of momentum in any any self-contained system is invariable. its Thus. there has been a violation of the Third Law of Motion. on this view. before the pulse reached the charged body there was momentum in the pulse. Third Law of Motion. view has been changed. if we suppose that momentum is necessarily confined to what we over it.. after the pulse passed over the body there was some momentum in the body and a smaller amount in the pulse. if we recog- momentum in the electric for. take the case of a charged body at rest struck by an electric pulse. are apparant thus. Action and Reaction are To the equal and opposite. the charged body when exposed to the electric force in the pulse acquires velocity violations of this principle and momentum. momentum in the charged body. according to Newton's tem. so that when the pulse has passed momentum is not what it was origiif we confine our attention to the nally. ÆTHE RF ORCE . i.

Then. is Many parallel to you will notice that the momentum what is known as Poynting's vector is the vector whose direction gives the direction in which energy flowing through the field. We now proceed to consider this momentum more in detail. then the momentum 0. volume the is equal to NB per unit sin the direction of momentum of being at right angles to the mag- netic induction and also to the Faraday tubes.LINES OF FORCE the loss of 25 momentum in the pulse being equal to the gain of momentum by the body. Let us begin with the simplest. the angle between the induction and the Faraday tubes. Moment of Momentum Due to cm Electrified Point and a Magnetic Pole of To familiarize momentum let ourselves with this distribution us consider some simple cases in detail. 7. B the pole. let A^ Fig. I have in my " Recent Researches on Electricity and Magnetism" calculated the amount of momentum at any point in the electric field. since the momen- ÆTHE RF ORCE . and have shown that if N is the number of Faraday tubes passing through a unit area drawn the magnetic at right angles to their direction. that of an electrified point and a magnetic pole. B induction. be the point.

the direction of rotation being everywhere the same. but since the system is spinning momentum round B. there will be a finite moment of momentum round B. is as direction spin- goes. Calculating the value A A of this from the expression for the momentum given above. we see that the momentum B we draw be perpendicular to the plane thus. Faraday tubes and also to induction. and whose centres A lie along that line. throughout the It is equivalent evident that the resultant in any direction is zero. that possessed by a top ning around A B. ABP any point coincides with the direction of the momentum at that point. these lines will form tion at a series of circles whose planes are perpendicular to the line 12. Let us now find what this distribution of field is momentum to. This distribution far as of momentum. a series of lines such that their direc.26 ELECTRICITY AND MATTER turn at any point direction of the P is at right angles to A P. we em obtain the very simple expression as the value of the moment of momentum about m e being the charge on the point and the strength of the pole. the the magnetic will if P. By means of this y AB ÆTHE RF ORCE .

moment of momentum angles to 8 em AB with an axis at right Let in the plane of the paper. the angle A is A . return to the system of the point and pole. \ \ and field that the change in turn of the field must be constant. but the whole moment of momentum of the system comprising point. ÆTHE RF ORCE . but ifc/ its axis is along A! IB instead of A B. / be the change in the momentum of A. - be equal and opposite to that gained by the If is pole. the change in the sin 0. The moment of momentum of the field \ moment ^ \ has thus changed. pole. since the whole momentum is zero. ^ momentum is still em.LINES OF FORCE 27 expression we can at once find the moment of momentum of any distribution of electrified points and magnetic poles. so the moment of momen- \ must be accompanied \ j by an equal and opposite change in the moment of momentum of the pole and FlGt 8 The momentum gained by the point must point. this conception of the momentum of the system leads directly to the evaluation of the force acting To on a moving pole. fied electric For suppose that of charge or a moving magnetic in the time 8 t the electrito point were to move from A A! the '.

The direction of the force acting on the point is at right angles to its velocity and magnetic force. it ÆTHE RF ORCE . I then 8 / and 8 axis 1 must be is equivalent to angles to a couple whose in is at right AB whose moment the plane of the paper. or . The value we have found for is the ordinary also to the F expression for the mechanical force acting on a moving charged particle in a magnetic field. There is an equal and opposite force acting on the magnetic pole. rate of increase of the t F perpenF being the -r' sup- g2 momentum. or the point is acted on by a of force equal to e multiplied by the component the magnetic force at right angles to the direction of motion. and must be at sin 6.2g 8 ELECTRICITY AND MATTER that of B. . If v is the A. Thus 8 e m I of the paper right angles to the plane and emAA'sm<f> Where velocity of </> is the angle f B A A'. We thus _ get Jb = A rea . A A s* v 8 1 and we get T_ e m v sm AB* This change in the momentum may be posed due to the action of a force dicular to the plane of the paper.

the line joining moment of momentum of the system will remain unchanged and there will not be any forces acting either on the pole or the point. The relative forces called into play are due to the if motion of the pole and point. Suppose the instrument is spinning with AB ÆTHE RF ORCE . The of pole distribution of momentum in the system some respects to about the line A B. We that in a top spinning can illustrate the forces acting on a moving elecand point is similar in trified body by the behavior of such a top. 9 represent about the axis A a balanced gyroscope spinning B. <f>. where H's the strength of the magnetic field.LINES OF FORCE 29 is may be written as ev Hsm <. direction. let the ball at represent A the electrified point. Thus. that at the magnetic pole. saying that when well-known electromotive force of induction due to motion in a magnetic field. The force acting on unit charge chanical force is therefore v Hsm <f>. This me- may be force thus regarded as arising from an electric v H sin and we may express the result by in body is force v moving If sin is <f> a charged a magnetic field an electric This force is the produced. the these are moving with the same them will not alter in velocity. let Fig.

the point A merely move in which horizontally forward in the direction is pushed. 9. point would do way.30 ELECTRICITY AND MATTER horizontal. then if against AB it with a vertical rod I push will not horizontally. but will also move vertias a charged cally upward or downward. and if pole at B. just FIG. it if pushed forward in the same were acted upon by a magnetic Maxwell's Vector Potential There is momentum a very close connection between the arising from an electrified point and a ÆTHE RF ORCE .

ÆTHE RF ORCE . B. and let m be the strength of either A is simple calculation shows that case the axis of the resultant moment of momen- tum in the plane P A B at right angles to P O. AB O P. and the Vector Potential of that system. and momentum equal in magnitude and opposite in direction at O. and that the of of the moment momentum is equal magnitude to e. we can at once find that due to a charge e of pression electricity at let a point P. the negative pole of this magnet be at A. in. and a little magnet AB . a quantity which plays a very large part in Maxwell's Theory of Electricity. at right A B -m at P angles to the plane P A B is the vector The vector in at called by Maxwell the Vector Potential P due to the Magnet. the in this positive at pole. AB -Qjk at P directed another at right angles to the plane P A B. A is the middle point of B.LINES OF FORCE 31 magnetic system. From the ex- we have given for the moment of momentum due to a charged point and a magnetic pole. O being m. mi where < the angle AB is makes with This moment of momentum to that equivalent in direction to a and magnitude due momentum e.

however. as due to permanent magnets. the magnetic resultant field is When.32 ELECTRICITY AND MATTER Potential I. that the moment of moelec- mentum about any trified line passing through the simple particle A e calculation field is shows that the whole momentum in the equivalent to a momentum I at the electrified ÆTHE RF ORCE . e (Vector Potential at arises field the If entirely from electric magnetic currents instead of from permanent magnets. there is a moment of mo- mentum. We may evidently tential at field is system of magnets. the P momentum of a system consisting of an electrified point and the currents will differ in some of its features from the momentum when the magnetic field is case. Calling this Vector we see that the momentum due to the charge and the magnet is e I at P and a momenequivalent to a momentum tum e Tat the magnet. In the latter we have seen. but no resultant momentum. due to is entirely electric currents. the momentum momentum e I at P to- gether with momenta at each of the magnets equal to due to that magnet). it is easy to show that there a momentum. so that equivalent to a extend this to any complex if / is the Vector Poin the P of this system. but vanishes.

from the Third Law of Motion. field is entirely due to currents this is a representation of the momentum in the field if the magnetic field is partly due to magnets we have in addition to this momentum at other P momenta at these magnets the magnitude of the . the momentum of any self- Now the contained system must be constant. Thus. of the momentum in the momentum consists (1) field . momentum any particular magnet is e times the Vector Potential at due to that magnet. we see that changes in the momentum of the field must ÆTHE RF ORCE . circuits and (3) the momenta of the magnets or carrying the currents. The well-known expressions for the electroat P motive forces due to Electro-magnetic Induction follow at once from this result. If the magnetic complete . For. whether the magnetic field is due to per- manent magnets or to electric currents or partly to one and partly to the other. the momentum when an electrified point is placed in the field at P is is equivalent to a momentum e Tat P where I the Vector Potential at P. Since (1) is equivalent to a momentum e 1 at the electrified particle. (2) the momentum of the electrified point.LINES OF FORCE point 33 I being the Vector Potential at P due to the currents.

ÆTHE RF ORCE . to the axes of y. Mu -fe F\* constant. e G. the com- ponents parallel to these axes of the Vector Potential at P. As the momentum remains constant.. eH at parallel and the momentum of the P charged point at P has for u and F.e. z . y.34 ELECTRICITY AND MATTER be accompanied by changes in the momentum of be the mass of the electrified the particle. neous changes in MSu + eF= <*** -4* dt dt From this equation we see that the point with the charge behaves as if it were acted upon by a me. z of its velocity. Mty simulta- Mw. w the components parallel to the axes of #. F. v. i. components Mu. G. Let M particle. U. -j } see that there are electric parallel to y and z respec- These are the well-known expressions of the forces due to electro-magnetic induction. hence if 8w and F are 0. by an electric force equal to - In a similar forces -j } way we -. and we see that they are a direct consequence of the tively. u. a?. chanical force parallel to the axis of x and equal = to - e -=-. then the momentum of the field is equivalent to momenta eF.

LINES OF FORCE principle that action opposite. ÆTHE RF ORCE . No effects due to this state can be . 35 and reaction are equal and will Headers of Faraday's Experimental Researches remember that he is constantly referring to called the " what he Electrotonic State " . thus he regarded a wire traversed by an electric current as being in the Electrotonic State when in a magnetic field. This Electrotonic State of Faraday is just the momen- tum existing in the field. detected as long as the field remains constant it is when it is changing that it is operative.

by which. If the number through a unit area at to if /a is P is placed at right angles the N. can picture to ourselves the Let us begin by considering the case of the moving charged sphere. as the last lecture. common tion of the centre of the sphere the momentum is at a point P at right angles to each of these directions and so is at right angles to OP in the plane containing and the line of motion FlG - P ia of the centre of the sphere. of electric force are radial are circles having for a : The lines those of magnetic force axis the line of mo. magnetic induction at P of Faraday tubes passing P the magnetic permeability of the medium ÆTHE RF ORCE . showed in we state of such a field. I tion WISH in this chapter to consider the connecin the electric field I between the momentum and the Faraday tubes. is.CHAPTER IT ELECTRICAL AND BOUND MASS.

. When the length of the cylinder is veiy great compared with its breadth.e. a mass of the surrounding medium 4?r /x N* per unit volume. tube carries with it a finite mass of water. i. We behave veiy much as suppose when moving behave and narrow cylinders long if these water moving endwise. and is in the direction of the ity of the component of the veloc- Now this Faraday tubes at right angles to their is exactly the momentum which if would be produced the tubes were to carry with them. carry very little water along with them. each unit length of the allel to their length. the velocity of TrpNv direction v being the sphere and of the angle O P makes with the motion of the sphere. surrounding the sphere. or ^n^N*v sin 0. TrpNv sin 0. at right angles to their axis. while when they move sideways. parthrough in fact the tubes to . when they move at right angles to their length. the tubes posno mass themselves and not carrying any sessing of the medium with them when they glide equal to through it parallel to their own length. the mass of water carried by it when moving endwise may be neglected in com- ÆTHE RF ORCE . momentum the rule given on page 25 the in unit volume of the medium at is By P NX length..e.ELECTRICAL MASS 37 sin 0. i.

gg ELECTRICITY AND MATTER it parison with that carried by mass no had tube if the ways . it possesses in virtue would have mass for sideways but none for end- wise motion. capacity thus.j\^ where N\& of thus. hence thus E is equal to the kinetic energy possessed veloc- by the bound mass when moving with the of the ity of light. but ~~j^ V* where V is the velocity with which light travels through the medium. M the mass of the bound ether in that volume. when moving side- beyond that which it of the water it displaces. We shall call the mass 4?r /i N z carried tubes in unit volume the mass of the by the bound ether. It is a very suggestive fact that the electrostatic energy E in unit volume is proportional to 2 . the amount number of Faraday tubes bound mass per unit length of ÆTHE RF ORCE . : ==E = 27rJV where of K is medium while M = specific inductive 4?r /u. This can easily be proved as follows the . 47r/u. the N*. The mass 2 bound ether the in unit volume is .

but also upon the number of a and velocity of the Faraday tubes in its neighborhood. however. We have many analogies to this in the case of dynamical systems . thus. placed in a mass of liquid whose venot disturbed by the vortex column. would m ÆTHE RF ORCE . We have seen that this is proportional to the tension in each tube.ELECTRICAL MASS each Farkday tube is 4?r pN. Since the mass of ether imprisoned by a Faraday tube is proportional to the number of Faraday N tubes in unit volume. in the case of a number of cylinders with their axes parallel. The following hydro-dynamical system is one by which we may illustrate the fact that the bound mass is proportional to the square of the number of Faraday tubes per unit volume. Suppose we have a cylindrical vortex column of if strength locity. tension being. always proportional to the mass per unit length of the string. mov- ing about in an incompressible liquid. so that we may stretched strings of variable mass and tension regard the Faraday tubes as tightly the . the momen- tum and of any cylinder depends upon the positions velocities of the cylinders in its neighborhood. we see that the mass and momentum Faraday tube depend not merely upon the configuration and velocity of the tube under consideration.

Thus. where A such a case are represented is the section of the vortex FIG. the column will imprison a m mass of liquid equal to that enthe by largest of the closed lines of flow. We see that some column of these lines in the neighborhood of the are closed curves. The lines of flow in in Fig. 11. Since the liquid does not cross the lines of flow. we can easily show that the mass of liquid imprisoned ÆTHE RF ORCE . is the strength of the vortex column and a the velocity of the undisturbed flow of the liquid. the liquid inside a closed curve will always umn and closed If remain in the neighborhood of the colwill move with it. and at right angles to the axis of the vortex column. column whose axis is supposed to be at right an- gles to the plane of the paper.40 ELECTRICITY AND MATTER be constant both in magnitude and direction. 11.

These tubes. carry with them an appreciable portion of the ether through which they move. as if free to twist in any direction. if as proportional to the number of tubes in unit area. Affective of Velocity on I will tJie Sound Mass now consider another consequence of the idea that the mass of a charged particle arises from the mass of ether bound by the Faraday tube as- sociated with the charge. set itself broadside to the direction of motion. through the fluid withLet us consider how a long. but will.ELECTRICAL MASS 41 by the column is proportional to ~. when they move at right angles to their length. setting itself so as to carry with it as much of the fluid through which it is moving as possible. while when they move to their length. would behave when moving through a liquid. the system illustrates Faraday the connection between the bound mass and the we take m strength of the electric field. you might expect at first sight. Many illustra- tions of this principle could be given. they glide parallel out setting it in motion. will not. shaped like a Faraday tube. one very ÆTHE RF ORCE . move point foremost. Such a body. on the contrary. CL Thus. narrow cylinder.

neither are they equally distributed. ÆTHE RF ORCE .e. the plane at right angles to the direction of motion of the sphere. so that if they were crowded into the equatorial region the pressure there would be greater than that near the pole. that the Faraday tubes repel each other. ELECTRICITY AND MATTER is that falling leaves do not fall edge but flutter down with their planes more or less horizontal. so that if were the only thing to be considered the Faraday tubes would be forced up into the i. we see that the Faraday tubes attached to the to the direction of this principle all sphere will tend to set themselves at right angles motion of the sphere. The a com- actual distribution of the Faraday tubes is promise between these extremes. If we apply this principle to the charged sphere.. all They are not crowded into the equatorial plane. This would tend to thrust the Faraday tubes back into the position in which they are equally distributed all over the sphere. equatorial plane. the density of the tubes in these regions increasing with the speed with which the charge is moving.42 familiar one first. for they are more in the equatorial regions than in the others the excess of . however. We be moving at right angles must remember. for in this position they would all to their lengths.

has been shown (see Heaviside. where V is the velocvelocity ity of light through the medium and v the that it is of the charged body. Phil. of the sphere at i." p. Mag. 1889. in such a way that the projection of the tube on this plane remains the same as for the uniform distribution of tubes. From this result we see only when the velocity of the charged body is comparable with the velocity of light that the change in distribution of the Faraday tubes due to the motion of the body becomes appreciable.. the plane through the centre right angles to its direction of motion.ELECTRICAL MASS 43 When it a Faraday tube is in the equatorial region it is imprisons more of the ether than when near the poles. In " Recent Researches on Electricity and Magnetism. 1 calculated the momentum I. so that the displacement of the Faraday tubes from the pole to the equator will increase the amount of ether imprisoned by the tubes. April. 19) that the effect of the motion of the sphere is to displace each Faraday tube toward the equatorial plane. but that the distance of every point in the tube from the equatorial plane proportion of is reduced in the V V*-v* to V. "Recent Researches." p. in the ÆTHE RF ORCE . 21..e. It and therefore the mass of the body.

and thus quence of the charge by we see from equation (1) that for velocities of the charged body comparable with that of light the mass of the body will increase with the velocity. where and 6 as before v and V are respectively = v j^.(!) the velight.. particles having masses far smaller than the mass of any known atom or molecule are shot out from radium with velocities ap- proaching in some cases to that of light. It is evident from equation (1) that to detect the influ- ence of velocity on mass we must use exceed- ingly small particles moving with very high velocities. and showed that the value of /is given by the following expression : cos 20)1 . increased in conse- . Now. is locities of the particle is and the velocity of given by the equation sm The mass 6 of the sphere . and the ratio of the electric charge to the mass for parti- ÆTHE RF ORCE ..44 ELECTRICITY AND MATTKK a sphere of radius a. having its space surrounding at the centre moving charged body.

that the mass increases with the velocity. if we suppose the charge to remain constant.59 2.83 2.10 1X 2. indicating. If then we find that the mass varies appreciably .77 2. results give us the Kauf mann's means of comparing the part of the mass due to the electric charge with the part the second independent of the electrification part of the mass is independent of the velocity. with the results shown in the following table first the column contains the values of the velocities of the particle expressed in centimetres per sec- ond.975 1. .17 1.31 We see from these values that the value of m di- minishes as the velocity increases.62 .36 . we infer that the part of the ÆTHE RF ORCE . with the velocity.ELECTRICAL MASS cles of this 45 kind has lately been made the subject of a very interesting investigation by Kaufmann.48 2. the second column the value of the fraction where particle : e is the charge and m the 10-7 mass of the v X 10.72 .

Kaufmann.46 ELECTRICITY AND MATTKI! mass due to the charge must be appreciable in comparison with that independent of it. for the effect on a charged sphere for example is not quite the same as that on a charged ellipsoid but having made the assumption and calculated the theoretical . C. F. To calculate the effect of velocity on the electrified mass of an system we must make some assump- tion as to the nature of the system. then if M M v. came to the conclusion that when the particle was moving slowly the " electrical mass " was ÆTHE RF ORCE . it is easy to deduce the ratio of the part of the mass independent of the charge to that part which at any velocity de- pends upon the charge. vi 1 are the observed masses at the velocities v and v respectively and M the part of the mass independent of charge. then two equations from which and m can be determined. effect of the velocity on the mass. Searle. on the assumption that the charged body behaved like a metal sphere. Suppose that the part of the mass due to electrification is at a velocity v equal to f(v) where f(v) is a known function m of v. the distribution of the lines of force of which M when moving has been determined by G.

ful to point out that this fraction He was care- depends upon it is the assumption we make as to the nature of the moving body. to the mass of the same by rawhen at or moving slowly. as. centre at the charged point on this supposition is the part of the mass due to the charge of in equation (1) the value calcu- on page 44.ELECTRICAL MASS 47 about one-fourth of the whole mass. and I prefer the simpler assumption that the distribu- tion of the lines of force round these particles is the same as that of the lines due to a charged point. I have lated from this expression the ratio of the masses of the rapidly moving particles given out particles dium rest. provided field outside we confine our attention to the its a small sphere of radius a having . on the assumption that the ÆTHE RF ORCE . evidently regarded as the most probable In the present state of our knowledge of the constitution of matter. I do not think anything is gained by attributing to the small negatively charged bodies shot out by radium and other bodies the property of metallic conductivity. and that with other assumptions his experiments might show that the whole mass was electrical. for example. whether spheri- cal or ellipsoidal. which he result. insulating or conducting.

TABLE . the number of times the mass of a particle moving with this velocity exceeds the mass of the same particle when at rest. the first column of which contains the values of v. the velocities of the particles . the second p.4g ELECTRICITY AND MATTER to th^ whole of the mass is due ratio as charge and have com- pared these results with the values of the same determined by Kaufmann's experiments. determined by equation (1) in his experiments. These results are given in Table (II). the third column p\ the value of this quantity found by Kaufmann ÆTHE RF ORCE .

elec- trostatic units. me and have been deteris mined the result of these determinations T that 5= e 10- and e = 1. the particle is as it mass of the has no definite limit. and In consequence. a length very small in comparison with 8 the value 10" c m. given by the equation m= -$ In a subsequent lecture I e will explain how the values of . which is usually taken as a good approximation to the dimensions of a molecule. were diffused through space. S. expression for 10~ 14 m Substituting these values in the we find that a is about 5 X cm. m due to the charge is e when the particle moving slowly. As these tubes stretch out to an infinite distance.ELECTRICAL MASS confine our attention to the part of the field is 49 which on the outside a sphere of radius a concentric with the charge. G. We have regarded the mass in this case as due to the mass of ether carried along by the Faraday tubes associated with the charge. however. of the very small size of the particle and the fact that the mass of ether carried by the tubes (being proportional to the square of the density of the Faraday tubes) varies inversely as the fourth ÆTHE RF ORCE . the mass particle is.2 X lO'80 in C.

50 ELECTRICITY AND MATTER particle. and. power of the distance from the we find by a simple calculation that all but the most insig- nificant fraction of mass is confined to a distance from the particle which is very small indeed compared with the dimensions ordinarily ascribed to atoms. On this of the mass of view of the constitution of matter. I that the hope atoms of the various elements are collections of is a later lecture positive and negative charges held together mainly by their electric attractions. In any system containing electrified bodies a part of the mass of the system will consist of the mass of the ether carried along by the Faraday Now tubes associated with the electrification. atom between the positively The view is I wish to put before you that it is not merely a arises mass of a body which in this ÆTHE RF ORCE . that the negatively electrified particles in the atom (corpuscles I have termed them) are identical with those small negatively electrified particles whose properties we have been discussing. part any body would be the mass of the ether dragged along stretching across the by the Faraday tubes and negatively part of the electrified constituents. moreover. one view of the constitution of matter to discuss in a view.

ELECTRICAL MASS 51 is way. and of the ether. constant to a very high degree of approximation. but that the whole mass of any body the mass of ether surrounding the carried just is body which Faraday tubes associated along by with the atoms of the bodj^. as observation and experiment have shown. and thus that electrified bodies mass instead of being. requires the density of the ether to be immensely greater than that of any known substance. In fact. These objections do not. apply to such a case as that contemplated in the preceding theory. in comparison various members of the system of ÆTHE RF ORCE . where the dimensions of one set of the electrified bodies the negative ones are excessively small with the distances separating the electrified bodies. all all momentum. kinetic energy This view. should vary with changes in the physical or chemical state of the body. It might be objected that since the mass has to be carried along by the Faraday tubes and since the disposition of these depends upon the relative position of the electrified bodies. should be said. kinetic energy. the mass of a collection of a number of positively and negatively would be constantly changing with the positions of these bodies. momentum it of the ether. however. that all mass the is mass of the ether.

the and charge. amount depending only on their size Thus. unless we alter the number or character of the corpuscles. the changes occurring in the mass through any alteration in their rela- be quite insignificant in comparison with the mass of the body.52 ELECTRICITY AND MATTER concentration of the When this is the case the lines of force corpuscles of the is on the small negative bodies the so great that practically the whole is bound ether localized around these bodies. tive positions will ÆTHE RF ORCE .

the tube in fact having very considerable analogy with a stretched string. Any disturbance communicated to one end of the tube will therefore travel along it with a constant and finite velocity . and consider what if we suddenly stop the point. they are also in a state of tension.CHAPTER III EFFECTS DUE TO ACCELERATION OF THE FARADAY TUBES Rontgen Rays and Light WE we is have considered the behavior of the lines of force when at rest and when moving uniformly. the tension at any point being proportional to the mass per unit length. The Faraday tubes associated with the sphere have must happen inertia. shall in this chapter consider the phenomena lines which result when the state of motion of the changing. moving so slowly that the lines of force are uniformly distributed around it. Suppose we have a tightly stretched vertical string moving uniformly. from ÆTHE RF ORCE . Let us begin with the case of a moving charged point.

12. ELECTRICITY AND MATTER and that we suddenly stop one end. then when a time. The shape of the string at successive if intervals will be as shown in Fig. what will happen to the string ? The end will come to rest at once. if V is the velocity with which a disturb- ance travels along the string. A and each part of continue to the string will in virtue of its inertia move as if nothing until the disturbance starting had happened to the end from reaches A it. the horizontal portion increasing as from the fixed end increases. A. 12. t. A Thus. the parts of the string at a greater distance than will be unaffected Vt from A by the stoppage. its distance ÆTHE RF ORCE . has elapsed after the stoppage of A.54 right to left. the length of j A Fio. but the forces called into play travel at a finite rate. and will have the position and velocity they would have had the string had continued to move uniformly forward.

while the portion P'Q' outside the outer sphere will be in the position it would have occupied if the particle had not been reduced to rest. these tubes will be in the position they would have occupied if they had moved forward with the velocity they possessed at the moment inner sphere. if O ' is the position the particle would have occupied if it had not been stopped.RONTGEN RAYS AND LIGHT 55 Let us now return to the case of the moving charged particle which we shall suppose suddenly brought to rest. describe with the charged particle as centre spheres. Thus.e. To tubes after a time the time occupied by the stoppage find the configuration of the Faraday t has elapsed since the beginning of the process of bringing the charged particle to rest. the other T). i. 13) the tube hence at . since two the radius V(t no disturbance can have reached the Faraday tubes situated outside the outer sphere. then.. P'Q' will be a straight line pass- ÆTHE RF ORCE . the time the portion of this tube inside the inner sphere will occupy the position OP. while inside the since the disturbance has passed over the tubes. they will be in their final positions. OPQ . the particle was stopped. consider a tube which. being T. when the particle was stopped was along the this will line be the t final position of (Fig. one having the radius Vt.

as the following calculation will ELECTRICITY '. has now in the component. to preserve its continuity the tube must bend round in the shell between the ing through two spheres. and thus be distorted into the shape Thus. and this tangential component implies a tangential electric force. The stoppage of the particle thus produces a radical change in the electric field ticle. to electric and magnetic forces much than those greater existing in the field when the particle was moving steadily. due to the par- and gives rise. the tube which before the stop- FIG. AND MATTER Thus. 13. page of the particle was shell a tangential radial. T is ÆTHE RF ORCE . If is we suppose that the thickness 8 of the shell so small that the portion of the Faraday tube then if inside it may be regarded as straight. OPP'Q'.

evsinO --^ ro . .RONTGEN RAYS AND LIGHT the tangential electric force inside the pulse. since its P magnitude is given by the equation. ev = sin TTS~- The force tangential Faraday tubes moving forward with the velocity Fwill produce at P a magnetic be at right If equal to V T. ticle t the time which has elapsed since the par. . was stopped since R = -y light. " . if r = OP. and P =Vt f . this force will angles to the plane of the paper and in the opposite direction to the magnetic force existing at before the stoppage of the particle . previously r to existing in the proportion of Thus. > it exceeds the magnetic force f ev sin -z S. we have 57 R T Where v is P'R 00' sin vt sin the velocity with which the particle was moving before it was stopped. the ÆTHE RF ORCE . d the angle OP makes with the direction of motion of the particle. the radial force. where Fis the velocity of /TT we have.

gg ELECTRICITY AND MATTER of the particle is pulse produced by the stoppage the seat of intense electric and magnetic forces which diminish inversely as the distance from the charged particle.e. It is easy to show that the momentum in the ÆTHE RF ORCE . whereas the forces before the as the particle was stopped diminished inversely square of the distance . of energy thus radiated depends i. if the particle with which is the particle the place where the cathode rays were stopped. upon is stopped instantaneously the whole energy in the field will be absorbed in the pulse and radiated away. The amount upon the S.. the thickness of the pulse. the remainder will appear as heat at abruptness stopped. The energy to be equal to in the pulse can easily be shown 2 ^V 8 ' 3 this energy is radiated outward into space. if it is stopped gradually only a fraction of the energy will be radiated into space. this pulse travelling out- ward with the velocity of light constitutes in my opinion the Rontgen rays which are produced when the negatively electrified particles which form the cathode rays are suddenly stopped by striking against a solid obstacle.

Let us take. two spheres having having their centres at the particle radii respectively equal to Vt and V (t ÆTHE RF ORCE . after the particle is stopped is zero. particle light. momentum in the field outside the pulse as there is no momentum in the space through which the has the whole momentum in the field pulse passed. the case when the was initially moving with the velocity of the rule stated on page 43 shows that before the stoppage the Faraday tubes were all congregated in the equatorial plane of the To find the configuration of particle. in that case The tubes would at a distance have been in a plane Vt in front of the particle. for example. if the particle had not been stopped. will stoppage of the distributed. the Faraday tubes after a time t we proceed as before by finding the configuration at that time moving of the tubes. stopping a charged particle whenever the distribution of the Faraday tubes in the state of steady motion has been determined. Draw and T). the same give us the effect of the Faraday tubes before the pulse principles. case The preceding investigation only applies to the when the particle was moving so slowly that were uniformly however.RONTGEN RAYS AND LIGHT pulse at any instant is 59 equal and opposite to the .

instead being . one a pulse propa- gated in the direction in which the particle was it mov- ing before was stopped. We thus have in this case two plane pulses.. outside tion of the tubes will be the same as if the particle the plane at the distance ticle. e. stopped. The PIG. preceding ap- method can be plied to the of case when if the charged its particle. 14. the other a X spherical pulse travelling outward in all directions. has velocity altered in any way thus.60 ELECTRICITY is AND MATTER stopping the where r the time occupied in the outer sphere the configuraparticle. 14. the tubes will be Vt in front of the parand this plane will touch the outer sphere. hence to preserve continuity these tubes must run round in the shell to join the sphere as in Fig. i. Inside the inner sphere the Faraday tubes will be uniformly distributed. had not been stopped. the velocity v of the particle instead of being ÆTHE RF ORCE .

At* but g7 is where /is the acceleration of the particle. Thus. hence we have e _ '' VTt~ equal to /. that it will give rise to a pulse in which the magnetic force If is given by the equation e&v and the tangential sin electric force e . T by ~_ Now A v sin the thickness 8 of the pulse is the space passed over by a wave of light during the time the velocity of the particle is changing. hence if 8 1 is the time required to produce the change Av in the velocity 8 TT = F8 sin 1. as on page 57. we can show. hence we have ._ j sin T_ e ^sin ~ These equations show that a charged particle whose motion is being accelerated produces a pulse of electric and magnetic forces in which the forces vary inversely as the distance from the particle. if a charged body were made to vibrate in ÆTHE RF ORCE . e e Av T_ Av 87 sin ~~ T* ~T~ .RONTGEN RAYS AND LIGHT reduced to zero is 51 merely diminished by A v.

are led to take the we of light as the following extracts same view of the propagation from the paper. in fact. Faraday says. " The view which I am so bold to put forward considers therefore radiations as a high species of vibration in the lines of force to connect particles which are known and also masses together.g2 ELECTRICITY AND MATTER such a way that its acceleration went through waves of electric and periodic changes. periodic out from the charged travel would force magnetic body. The method we have been investigating. We from the propagation of transverse tremors along the tightly stretched Faraday tubes." show to have been taken by Faraday himself. in which we consider the effect produced on the configura- tion of the Faraday tubes by changes in the moof tion of the body. be light waves. on the Electromag- Theory of Light. "Thoughts on Kay Vibrations. provided the periodic changes in the acceleration of the charged body took place with sufficient rapidity." This view of light as due to the tremors in tightly stretched Faraday tubes raises a question ÆTHE RF ORCE . affords a very simple way picturing to ourselves the processes going on dur- ing the propagation of a wave of light through have regarded these as arising the ether. netic These waves would.

RONTGEN RAYS AND LIGHT which I 63 have not seen noticed. pass ÆTHE RF ORCE . We have seen that the method of propagation and constitution of a Rontgen ray is the same as in a light wave. will be represented by a series of bright specks on a dark ground. and the front of the wave. Such a view of the constitution of a light wave would explain a phenomenon which has always struck me as being very remarkable and difficult to reconcile with the view that a light wave. They are rather to be looked upon if as discrete threads embedded in a continuous ether. as it were. instead of being. then on the view we have taken of a wave of light the wave itself must have a structure. uniformly illuminated. does not possess a structure. gen rays will apply also The phenomenon are able to in question is this: Rontgen rays very long distances through as and they pass through the gas they gases. The Faraday tubes stretching through the ether cannot be regarded as entirely filling it. so that any general consideration about structure in Rontto light waves. or rather in this case a Rbntgen ray. the bright specks corresponding to the places where the Faraday tubes cut the wave front. giving to the latter a fibrous structure . but this is the case.

ionize tive


and negative ions

up the molecules into posithe number of molecules

so split

up is, however, an exceedingly small fracless than one-billionth, even for strong rays, of
of molecules in the gas.



Now, if


conditions in the front of the

wave are uniform,

the molecules of the gas are exposed to the same conditions how is it then that so small a propor;

tion of


are split

up ?


might be argued that

those split


are in

special condition

they possess, for example, an

amount of


the average kinetic of the molecules gas that, in accordenergy of the ance with Maxwell's Law of Distribution of

energy so

much exceeding

Kinetic energy, their number would be exceedingly small in comparison with the whole number of

molecules of the gas



if this

were the case the

same law of distribution shows that the number
in this abnormal condition

would increase very with the rapidly temperature, so that the ionization produced

by the Rontgen

rays ought to


crease very rapidly as the temperature increases.

Recent experiments made by Mr. McClung in the Cavendish Laboratory show that no appreciable
increase in the ionization is produced



the temperature of a gas from


to 200




whereas the number of molecules possessing an

abnormal amount of kinetic energy would be
enormously increased by this
rise in



difficulty in explaining the small ionization is


instead of supposing the front of the
to be uniform,

Rontgen ray
sists of

we suppose that it


intensity separated by considerable intervals where the intensity is very small, for in this case all the molecules in the field,

specks of great

and probably even


parts of the same

molecule, are not exposed to the

same conditions,

and the case becomes analogous to a swarm of cathode rays passing through the gas, in which
case the


of molecules which get into col-

with the rays

may be

a very small fraction

of the whole


of molecules.


return, however, to the case of the charged


whose motion



we have

seen that from the particle electric and magnetic
forces start

and travel out radially with the veof light, both the radial and magnetic forces

being at right, angles to the direction in which

but since (see page 25) they are travelling each unit volume of the electro-magnetic field has an amount of momentum equal to the product of

the density of the Faraday tube and the magnetic


force, the

direction of the

momentum being


to both these quantities, there will be right angles the wave due to the acceleration of the charged

and indeed in any

electric or light



in the direction of propagation of the


wave of

any such wave, for example a absorbed by the substance through

it is

passing, the


in the


be communicated
will, therefore,

to the absorbing substance,


in the direction

experience a force tending to the light is travelling.

when light

normally on a blackened abwill repel that substance.

sorbing substance,

This repulsion resulting from radiation was shown by Maxwell to be a consequence of the Electro-

magnetic Theory of Light
tected and measured
beautiful experiments,



has lately been de-

by Lebedew by some most which have been confirmed

and extended by Nichols and Hull.


pressure experienced

by the absorbing


stance will be proportional to its area, while the

weight of the substance is proportional to its volume. Thus, if we halve the linear dimensions we
reduce the weight to one-eighth while we only reduce the pressure of radiation to one-quarter thus, by sufficiently reducing the size of the absorbing


and all protected from external radiation. so that the refracting substance must have to is momentum communicated to It is easy show that even when the incidence oblique the of the light to the momentum communicated ÆTHE RF ORCE . while all bodies smaller than this would be repelled from the sun. the course of the light and therefore the direction of momentum it. in diameter the repulsion due to the radiation from the sun would just balance the sun's attraction. overpower their gravitational that at- so the spheres will repel each Again. like weight.RONTGEN RAYS AND LIGHT 67 body we must arrive at a stage when the forces due to radiation exceed those which. the re- pulsion due to the radiation emitted from the spheres will traction other. Arrhenius has for an shown that 6 opaque sphere of unit density lO" cm. knowing the radiation from the sun. that in if phenomena connected with Poynting-has recently shown two spheres of unit density about 39 cm. and he has applied this principle to explain the the tails of comets. diameter are at the temperature of 27 C. are proportional to the volume of the substance. when light is refracted and reflected at a transparent surface. is changed. intensity of the On this principle.

for example. we take into particle when if ac- Thus. must exist in certain cases .gg ELECTRICITY is AND MATTER surface. The waves radiate of electric and magnetic force which from an accelerated charge particle cany energy with them. the particle ÆTHE RF ORCE . The rate at which energy is radiating from o the particle can easily be e is the shown to be where charge on the particle. as been detected experimentally. so that the particle is constantly losing energy. so far as I know. and V the velocity of If count this loss of energy by the its motion is being accelerated. however. This energy is radiated into space. JT. Tangential due to light have not. we find some interesting results. These. such. for example. e starting a particle of rest is acted mass m and charge from upon by a constant electric force. substance normal to the refracting There are many interesting problems connected with the forces experienced by refracting prisms passing through them which will suggest themselves to you if you consider the changes in momentum experienced by the light when light is wave forces in its course through the prism./ light. when flected light incident obliquely is imperfectly re- from a metallic surface. its acceleration.

The " hard " rays correspond to . " thin pulses. the rate at which the e 9 its final acceleration. the particle were acted on force by a wave of electric e* which only took a time comparable with -y to pass over the particle. Thus. the acceleration of the partiis initially zero.RONTGEN RAYS AND LIGHT does not at once attain the acceleration e - 69 as it m would tion . and it is not until after the lapse of a time comparable with particle acquires -^ that the even an appreciable fraction of Thus. particle loses energy is during the time -y very if small compared with the ultimate rate. the " soft ones to thick ones a smaller proportion of the energy in the hard rays will be radiated away by the charged particles " so that " ÆTHE RF ORCE . the amount of energy radiated by the particle would be a very much smaller fraction of the energy in the wave than it would be if the particle took a time equal to a considerable e* multiple of 77^ to pass over the particle. cle if there were no loss of energy by radiaon the contrary. This has an important application in explaining the greater " " hard Rontgen rays than penetrating power of " of " soft ones.

radiation increases very Thus. we find that in this case the rate of radiation is proportional to the eighth power of the velocity. or to the fourth power of the energy. By ergy applying the law that the rate at which en- 1*V is 2 radiating is equal to ^ -y^. the rate of loss of energy by much more rapidly than the energy of the moving body. ÆTHE RF ORCE .70 ELECTRICITY AND MATTER " " over which they pass than in the case of the soft the case of a charged particle revolving in a circular orbit un- der an attractive force varying inversely as the square of the distance.

with their ten- mass of ether they carry along with them. Let us first consider the evidence given by the laws of the electrolysis of liquids. what may be called an atomic any charge being built up of a number : of finite individual charges. all equal to each other just as on the atomic theory of matter a quantity of hydrogen is built up all of a number of small par- ticles called atoms. in this chapter we shall discuss the nature of the charges of electricity which form the shall show beginnings and ends of these lines. We that there are strong reasons for supposing that these changes have structure . electric disturbances and with the propagation of along them .CHAPTER IV THE ATOMIC STRUCTURE OF ELECTRICITY HITHERTO we have been dealing properties of sion. each extremity of a Faraday tube will be the place from which a constant fixed num- ber of tubes start or at which they arrive. tricity is view of the structure of correct. Faraday ÆTHE RF ORCE . the chiefly with the the lines of force. If this the atoms being equal to elec- each other.

again. sodium. or three times the charge. of all We divalent ment carry twice the charge carried by the same number of ions of a univalent element. Thus. a tri- valent ion carries three times the charge of a univalent ion. number first of atoms coming up Let us consider monovalent elements. chlorine. he showed that of when the same deliver number up of atoms these substances their charges to the electrode. the the to positive electrode. such as hydrogen. the quantity of electricity communicated is the same whether the carriers in- are atoms of hydrogen. find again that the ions elements carry the same charge. in the case of the electrolysis of solutions the charges carried by the ions are either the charge on the hydrogen ion or twice that charge. chlorine. or sodium. and of posiity given up tive electricity given to the negative electrode. and so on. and so on . but that a number of ions of the divalent ele- the divalent elements. is proportional to the to the electrode. showing that each ion of a divalent element carries twice as much charge as the univalent ion . dicating that each atom of these elements carries Let us now go to the same charge of electricity. and ÆTHE RF ORCE .y2 ELECTRICITY AND MATTER electricity passes showed that when through a amount of negative electricliquid electrolyte.

" the conduction of electricity through gases. by exposure to Rontgen rays it remains in this state for a sufficiently long time after the rays have ceased to enable us to study its properties. lecture. we never meet with fractional parts of this charge. Let us consider for a moment a few of the propgaseous conduction. the evidence in favor of the atomic character of electricity is When we consider even stronger than it is in the case of conduction through liquids. We find that the conductivity out of the gas by the sending gas through a plug of cotton. is divided into definite elementary portions which behave like atoms of electricity. 73 The charges we meet with are always an integral multiple of the charge carried by the hy- drogen atom as " if . positive as well as negative. or through a water-trap. This very remarkable fact shows. Thus. we cannot avoid we the conclusion that electricity. erties of When a gas has been put into the conducting state say. the conductivity is we can filter due to something mixed with the gas which can ÆTHE RF ORCE .wool. that Helmholtz said in the Faraday accept the hypothesis that the elementary substances are composed of atoms.THE ATOMIC STRUCTURE OF ELECTRICITY so on. chiefly because tricity we know more about the passage of elec- through gases than through liquids.

where it can be meas- ured by an electrometer. is One method by which this can be done to enclose the gas between two parallel metal plates. e. by The lows. the conductivity arising from the motion of these particles in the electric field. the conductivity is taken out of the gas strong electric field. one of which is insulated. As this charge is equal ÆTHE RF ORCE . and these before they have time to combine with the negative particles will be driven against the insulated plate. We have at the Cavendish Laboraelectricity carried tory measured the charge of those particles.74 ELECTRICITY AND MATTER filtered be out of it. all the positive charge in the gas will be driven against the insulated plate. any time there are in the gas n of these charged positively and n charged negaif each of these carries an electric charge we can easily by electrical methods determine n e. when it is sent through a This result shows that the is constituent to which the conductivity of the gas due consists of charged particles. Thus. again. Now suppose we suddenly charge up the other plate positively to a very high potential. and the quantity of electricity of our sign present in the gas. this plate will now repel the positive particles in the gas. principle of the If at method first used is as fol- particles tively.

E. since there are . In dust-free get a Aitken showed. first and they are thus rendered visible. in sufficiently supersaturated damp air a is deposited on these charged particles. get their We number number indirectly as follows : suppose we have a of these particles in dust-free air in a closed vessel. when no charged particles are cloud Thus. the air being saturated with water vapor. when as the particles are surrounded by damp air cooled below the saturation point. however. step toward counting them. it is very difficult to fog when damp air is cooled. supersat- a fog will be deposited round these by a uration far less than that required to produce any appreciable effect present. C. far too small and too numerous to be counted directly. can. This is the The drops are. we can then we can to e n in this way easily determine n if shall be able to find Wilson that the charged particles act as nuclei round which small drops of water condense. no nuclei for the drops to condense around are charged particles in the dust-free air. suppose now that we produce a sudden ÆTHE RF ORCE .THE ATOMIC STRUCTURE OF ELECTRICITY 75 e : n we The method by which I determined n was founded on the discovery by devise a means of measuring e. if there however. T. however. air.

To find the size of a drop we make use of an investigation Thus.21 X 10 6 a2 . spheres fall through the In consequence of the viscosity of the air small bodies fall exceedingly slowly.76 ELECTRICITY AND MATTER the expansion of the air in the vessel. so that if we know the volume of one drop we can deduce the number of drops. ÆTHE RF ORCE . and the smaller they are the slower they fall. we know the volume of water in the form of drops. we can determine v we can determine the radius and hence the volume of the drop. produced we can calculate the cooling of the gas. this will cool air. it will be supersaturated with vapor. is the acceleration due to 981 gravity the coefficient of viscosity of air = .00018. and drops will be deposited round the charged partiNow if we know the amount of expansion cles. by Sir George Stokes on the rate at which small air. the velocity v with which through the air is falls given by the equation when g and thus p. and therefore the amount of water deposited. Stokes showed that if a is the radius of a it drop of water. = v hence if = 1.

found that 3. and is But v can easily be measured by observing the movement of the top of the cloud. S. units. In this way I found the volume of the drops. Experiments were made with air. G.c. As n e had been determined electrical measurements. a strong argument of electricity. in a of a gas at this temperature and pressure the is number of hydrogen ions in 1.c. hydrogen at the temperature 15C. or 3 X 10 10 electrostatic of through acidulated water liberates 1.23 2. that the passage of one electro-magnetic unit of electric charge. and it was found that the ions had the same charge in all these gases. the value of e could be deduced when n was known in this way I by . and pressure .4 its value 10 is X 10- Electrostatic C. of one atmosphere c. and carbonic acid.46 N. if there are N molecules c.23 c.THE ATOMIC STRUCTURE OF ELECTRICITY 77 evidently the velocity with which the cloud round the charged particle settles down. in favor of the atomic character We can compare the charge on the gaseous ion with that carried by the hydrogen ion in the electrolysis of solutions in the following way: We know units. hydrogen.c. and thence n the number of particles. ÆTHE RF ORCE .

or E= 1. however. . is equal to the charge on the Dr. the charge on the gas ion 3. H. in the kinetic theory of gases methods are ^Y=3. in good of the best of these deterwith some agreement minations. its near neighbor- X10 is. by quite a different method.7g ELECTRICITY AND MATTER if so that E is the charge on the hydrogen ion in the electrolyis of solution. or Avogadro's Constant. Wilson. 2.22 X 10 19 -f-JV: is Now.4 X10" 10 . T. and hence we conclude that the charge 3. A.6 The value 19 on the gaseous ion electrolytic ion. as it values obtained is sometimes called the by this with the assumptions made theory vary somewhat as to the nature of the molecule and the nature of the forces which one molecule exerts on another in hood.6X10 investigated for determining this quantity N. R. the charge on the gaseous ion will equal the charge on the electrolytic ion. obtained practically the same value for e as that given above. Wilson that it requires less supersaturation to deposit clouds from moist air on negative ions ÆTHE RF ORCE .46 NE= 3X 10 10 10 . Now. His method was founded on the discovery by C. of the Cavendish Laboratory. hence if e.

can be at once Townsend showed gaseous ion is that the charge on the equal to that on the ion of hydrogen coeffi- in ordinary electrolysis. by observing the rate at which the cloud falls we can. the plate will attract the cloud. and we can adjust the charge on the plate until the electric attraction just balances the weight of a drop. Thus. suppose we place above the cloud a positively electrified plate. as explained above. As is e is equal to the weight of the drop which e known. so that each drop in the cloud is negatively charged the supersaturation. and the drops. Now.THE ATOMIC STRUCTURE OF ELECTRICITY than it 79 does on positive. by suitably we can get the . determine the weight of each drop. hang stationary in the air. like Mahomet's coffin. Let us consider the cient of diffusion of it volume of ionized gas between two horizontal planes. if X is the electric force then the electric attraction on the e is drop is X Xe. choosing cloud deposited on the negative ions alone. determined. when the charge on the drop. and as we can measure X. by measuring the the gaseous ions and comwith the velocity acquired by the ion paring under a given electric force. and suppose that as long as we case of a keep in any horizontal layer the number of ions ÆTHE RF ORCE .

is the velocity when the electric ÆTHE RF ORCE . variation in the partial pressure due to the ions if this pressure is equal top. as Rutherford and Zeleny have done.8Q ELECTRICITY AND MATTER remains the same. the force acting on the ions in a unit volume . the velocities acquired by the ions in an electric field. n the number of ions of one sign in unit volume be the coefficient of if layer. but that the number varies as we pass from one layer to another. of this in one second pass D downward through unit area of the layer is dn ir*' so that the average velocity of the particles down- ward is l)dn n dx The force which sets the ions in motion is the . is T-> and the average force can find the velocity per ion is -r ndx - Now we which an ion acquires when acted upon by a known force by measuring. let x be the distance of a layer from the lower plane. They showed that this velocity is proportional to the force acting on the ion. then of ions which number the of the diffusion ions. so that if A.

A e and the velocity when the force 1 n ax ^ will therefore be i this velocity dp ^L.THE ATOMIC STRUCTURE OF ELECTRICITY force is 81 JTand when the therefore -==r-. force acting on the ion is the velocity for unit force will be is . In this way Townsend found ÆTHE RF ORCE .e. for all gases. X e. by knowing of D and a we can find the value that Ne. so that i. the number of mol- ecules in a cubic centimetre of gas at the atmos- P p P~W Ne was _ n and equation (1) gives us PA Thus. to be equal to D hence dn f n dx' we have dAdn P ~Tr~ --T) ~7 dx Xe */ dx "7 (l\ 1/ V Now if the ions behave like a perfect gas. at the This ratio is the same if same temperature. the number of ions per unit volume. n dx Xe' we have seen. however.. jV is Avogadro's pheric pressure constant. the pressure p bears a constant ratio to n.

oxygen. gas.g2 ELECTRICITY AND MATTER air. as in all cases of conduction a maximum through gases." electric field ap- which was not exceeded until the plied to the gas approached the intensity at which ÆTHE RF ORCE . attained value called the " saturation current. The current. these experiments show that eE. We have seen that E the charge on the hydrogen ion NE = 1. for after the current had reached a certain value but this increase did no further increase in the electromotive force produced any change in the current. or that the charge on the gaseous ion is equal to the charge carried by the hydrogen ion in the electrolysis of solutions.24 X 10 10 . the same in acid. and the hydrogen. The current through the vapor increased at first with the electromotive force used to drive it through the not go on indefinitely. The equality of these charges has also been proved in a very simple way by H. measured quanThis vapor got ionized and the mixture of air and vapor acquired very considerable conductivity. Wilson. a tity of the vapor of metallic salts. A.22 X 10 10 - Thus. who introduced per second into a volume of air at a very high temperature. and carbonic of his values if mean was is N e = 1.

we are led to the conception of a natural unit or electricity of tiples. found that the saturation current through the salt vapor was just equal to the current which if it passed through an aqueous solution of the salt would electrolyse in one second the same amount of salt as It is was fed per second into the hot air. e the charge on an ion.N" is this constant. Mass of the Carriers of Electricity We must now pass on to consider the nature of the systems which carry the charges. atom of gen is charges are integral muljust as the mass of a quantity of hydroan integral multiple of the mass of a all which hydrogen atom. jV= 3.22 10 10 and X we have seen that e = 3. whether we study the conduction of elec- tricity through liquids or through gases.THE ATOMIC STRUCTURE OF ELECTRICITY 33 Wilson sparks began to pass through the gas.fc X 10 19 . or of the way in which shape they act upon each other. then JV e 1. If . Thus.4 X l^ 10 so that . and in order to have the conditions as simple as possible ÆTHE RF ORCE . worth pointing out that this result gives us a method of determining Avogadro's Constant which is independent of any hypothesis as to the or size of molecules.

where impeded by gas. v by the equation mv may be t The ticle is velocity v of the particle : deter- mined by the following method Suppose the par- moving horizontally in the plane of the ÆTHE RF ORCE . so that the path of the particle will be that described by a body acted upon by a constant normal force. and that it is acted on by a uniform magnetic field at right angles to this plane. carrying a charge e. where His the magnetic force and particle. moving in the plane of the paper. always at right angles to the direction of motion of the particle.g4 let ELECTRICITY AND MATTER us begin with the case of a gas at a very low the motion of the particles is not pressure. It is easy to show is that this path a circle whose radius a is given . v the velocity of the to The direction of this force is in the plane of the paper at right Since the angles force is the path of the particle. We have seen that under these circumstances the particle will be acted upon by a mechanical force equal to He v. the velocity of the particle and therefore the magnitude of the force it act- ing upon will not alter. collisions with the molecules of the Let us suppose that we have a particle of mass m.

through a uniform magnetic field If. means of finding both v and ÆTHE RF ORCE . and determine the value of the electric force Equations (1) and (2) then give us the required to counteract the effect of the magnetic force. circle into we have methods of we can measure which it is tracing the motion the radius a of the bent by a constant magnetic force. When the two e forces are equal we have X Hence if H e v. or of the particle. tell We can when forces are absent. Now. since in this case the motion of the particle under the action of the electric and magnetic forces will be the same as when both these are equal. the particle will be acted upon if by a vertical force equal to X. at right angles to this plane. and adjust the value of X until the this two forces adjustment has been made. Let us arrange the direction of JTso that this force is in the opposite direction to that due to the magnet.THE ATOMIC STRUCTURE OF ELECTRICITY 35 paper. e H e v. a in addition to the magnetic force we apply vertical electric force cal this will exert a verti- mechanical force X on the moving particle.

gg ELECTRICITY AND MATTER Values of for Negatively Electrified Particles in Gases at Low Pressures The value of e m . (1) when exposed to ultra-violet light. gram. (2) when raised to the tempera- ture of incandescence. or whatever the it nature of the metal from which may be sup- posed to have proceeded. These experiments have led to the very remarkable result that the value of is 171 the same whatever the nature of the gas in particle which the may be found. and second. p In fact. has been determined in this way for the negatively electrified particles which form the cathode rays which are so conspicuous a a gas at low part of the electric discharge through pressures . the units being the centimetre. in every case in which the value of m has been determined for negatively electrified particles ties moving with veloci- than the velocity of light. As the value of for the hydro- ÆTHE RF ORCE . and also for the negatively electrified particles emitted by metals. it has been found to have the constant value about considerably less 10 T. and the charge being measured in electro- magnetic units.

and I have ÆTHE RF ORCE . The negative electric fluid. has thus a structure analogous to that of a gas. . I units of negative electricity. in Negative a gas at a low pressure. for the carriers of positive electri- positive This has been done by Wien for the electrification found in certain parts of the discharge in a vacuum tube." to use the old notation. we see that the mass of a carrier of the negative charge must be only about oneion is thousandth part of the mass of hydrogen atom the mass was for a long time regarded as the smallest mass able to have an independent existence. Carriers of Positive Electrification We can apply the same methods to determine fy the values of fication. " taking the place of the molecules. 4 gen ion in the electrolysis of liquids is only 10 and as we have seen the charge on the gaseous equal to that on the hydrogen ion in ordinary electrolysis. are the have proposed the name corpuscle for these These corpuscles same however the electrification may have arisen or electricity.THE ATOMIC STRUCTURE OF ELECTRICITY 37 . the corpuscles wherever they may be found. resembles a gaseous fluid with a corpuscular instead of a molecular structure.

cases the value of In many 4 . measured off by a hot The results of these measure- ments form a great contrast to those for the negative electrification. the electric fluid to "One be positive electricity we take it to be negative. as Franklin did. for for the positive charge. as has for the negative.CQ oo ELECTRICITY AND MATTER for the positive electrification given wire. just as of the it would if the carriers positive charge were the atoms of the elements which happened to be present when the positive electrification These results was produced. The "electric fluid" of Franklin cor- ÆTHE RF ORCE . is found never to have a 4 value greater than 10 the value it would have if . the carrier were the atom of hydrogen. it instead of having. the 7 constant high value 10 . m is very much less than 10 inis dicating that in these cases the positive charge carried by atoms having a greater mass than that of the hydrogen atom. The value of ^ varies with the nature of the electrodes and with the gas in the discharge tube." Instead of taking. lead us to a view of electrification to which has a striking resemblance Franklin's that of Fluid Theory of Electricity.

We " " than in fact know more about the electric fluid we know about such fluids as air or water. negative electrification being a collection of these corpuscles. The transference of electrification from one is place to another effected by the motion of cor- puscles from the place where is a gain of positive electrification to the place where there is a gain of negative. A positively electrified body is one that has lost some of its corpuscles. ÆTHE RF ORCE .TI1K ATOMIC STRUCTURE OF ELECTRICITY 9 responds to an assemblage of corpuscles. We have seen that the mass and charge of the corpuscles have been determined directly by experiment.

been advanced by more than one chemist. many That in fact the atoms of these substances We have something in common.CHAPTER V CONSTITUTION OF THE ATOM WE have seen that whether we produce the corpuscles by cathode rays. up of atoms of hydrogen weights of all if this were so the combining the elements would. or from incandescent metals. Thus Prout. and since the mass of the corpuscles is less than that of any known atom. are thus confronted with the idea that the atoms of the chemical elements are built up of simpler systems an idea which in various forms has . on the assump- ÆTHE RF ORCE . and whatever may be the metals or gases present we always get the same kind of corpuscles. by ultra-violet light. in 1815. put forward the view that the atoms of all the chemical elements are built . we see that the corpuscle must be a constituent of the atom of different substances. Since corpuscles similar in all respects may be obtained from very different agents and materials.

To avoid this discrep- suggested that the primordial atom not be the hydrogen atom.CONSTITUTION OF THE ATOM tion that there 91 was no loss of weight when the atoms of hydrogen combined to form the atom of some other element. The simple relations which exist between the combining weights of several of the elements having similar chemical properties. which shows that there of the elements a periodicity in the properties when they are arranged in the order of increasing atomic weights. all point to the conclusion that the atoms of the different elements have something in common. Further support was given to the idea of the complex nature of the atom by the discovery by Newlands and Mendeleeff of what is known is as the periodic law. for example. Further evidence in the same direction is afforded by the similarity in the structure of the spectra of elements in the same group in the periodic series. be integers a result not in ac. cordance with observation. but a smaller might ancy Dumas atom having only one-half or one-quarter of the mass of the hydrogen atom. a similarity which recent work on the existence in spectra of series of lines whose frequencies are connected by definite ÆTHE RF ORCE . the fact that the combin- ing weight of sodium is the arithmetic mean of those of lithium and potassium.

g2 ELECTRICITY AND MATTER much to emphasize indeed spectroscopic evidence alone has led Sir Norman Lockyer for a long time to numerical relations has done . of The phenomenon have to speak of which I shall later. although the model of the atom to which led we are by these considerations is very crude and im- perfect. ÆTHE RF ORCE . sider the bearing of the existence of corpuscles on the problem of the constitution of the atom and . radio-activity. If this is so then we must face the problem of the constitution of the if atom. and establish advocate the view that the elements are really compounds which can be dissociated when the circumstances are suitable. for there seems good reasons for believing that radio-activity is due to changes going on within the atoms of the radio-active substances. it may perhaps be of service by suggesting us with fur- lines of investigations likely to furnish ther information about the constitution of the atom. and see has in it we can imagine a model which re- the potentiality of explaining the markable properties shown by radio-active subIt may thus not be superfluous to constances. carries the argument still further.

CONSTITUTION OF THE ATOM 93 The Nature of the Unit from which the Atoms are Built Starting from the is Up that the atom hypothesis an aggregation of a number of simpler systems. We is have so seen that the cor- whose mass much less than that of the a constituent of the atom. For reasons which will appear later on. the end and an equal positive charge two ends being connected by lines of electric force which we suppose to have a material existence. carries a definite charge of negative electricity. Let us then take as our primordial system an with a negative corpuscle at one at the other. however. therefore be very The lines of force will much more condensed near the ÆTHE RF ORCE . systems. it is natural to atom. the corpuscle as a constituent of the primorregard is dial system. and since with any charge of electricity we always associate an equal charge of the opposite kind. we should expect the negative charge on the corpuscle to be associated with an equal charge of positive electricity. electrical doublet. we shall suppose that the volume over which the positive electricity is spread is very much larger than the volume of the corpuscle. let us consider what is the nature of one of these puscle. The corpuscle.

as the charge on the corpuscle and a 1013 its radius we have seen. will be very corpuscle than elsewhere. posed. ÆTHE RF ORCE . the practically arise mass of the system will from the mass of bound ether . This mass (see page 21) is for each corpuscle equal to . and aggregations of more than one system would be formed. the size of the corpuscle we regard as the mass much greater near the If. suppose we had a universe consisting of an immense number of these electrical doublets. just as the attractions of a lot of magnets would draw them together if they were free to move. them little together.94 ELECTRICITY AND MATTER other part of the system. thus the mass of the system will be practically independent of the position of its positive end. and corpuscle than at any therefore the quantity of ether bound by the lines of force. the mass of which of the system. and will be very approximately close to the corpuscle the mass of the corpuscles if alone in the field. Now which we regard as our primordial system if these were at rest their mutual attraction would draw . being about cm. where e is a. as we have supis pared with the size very small comof the volume occupied by the positive electrification.

the rel- two systems. 95 however. i. the tendency to separate due to their relative motion was not prevent them remaining together under their mutual attraction. come whenever they come so close together that they sensibly acceler- ate or retard each other's motion. since it electrical rying energy with them. Let us consider for a moment the way in which the kinetic energy of such an assemblage of units have seen (p. the ÆTHE RF ORCE . carThus. whenever the into collision. 68) that whenwould diminish. energy will be radiated away. although it may be a very long time..CONSTITUTION OF THE ATOM If. the individual systems were originally moving with considerable velocities. In this case the formation of aggregates would be postponed. units waves which radiate through space. There will thus be a steady loss of kinetic energy. until the kinetic energy of the units had fallen so low that when they came sufficient to into collision. when they came near enough to exercise appreciable attraction on each other. might be sufficient to carry the sysative velocity of tems apart in spite of their mutual attraction.e. the whole of which will not be absorbed by the surrounding units. and after a time. We ever the velocity of a charged body the is changing generates body is losing energy.

and exerting therefore a radial electric force proportional at an internal point to the distance from the centre.e. ELECTRICITY AND MATTER kinetic energy will fall to the value at which aggrewill begin . Let us picture to ourselves the aggregate as. like the ^Epinus atom of Lord Kelvin. moving about The number of corpusof units cles is the number which had gone to make up the aggregate. aggregates containing In considering the question of the further aggregation of these complex groups. To fix let us take the case shown in Fig. consisting of a sphere of uniform positive electrification. and the total negative electrification Fio. 15 on the corpuscles our ideas is equal to the positive tion electrifica- on the sphere. 15 of three corpuscles ÆTHE RF ORCE . upon the velocity of the centre of gravity. electri- and that the very much smaller negatively fied corpuscles are inside it. gation of the units into groups of these will later on be followed of two by the formation a larger number of units. but also upon the relative velocities of the corpuscles within the aggregate. we must remember that the possibility of aggregation will depend not merely upon the velocity of the aggregate as a whole.

are describing circular orbits round the centre of the sphere. in- Next suppose that the stead of being at rest. electrification of A simple is calculation shows that this will be the case when the distance of the corpuscle from the centre the sphere.CONSTITUTION OF THE ATOM 97 A. force will carry Their centrifugal centre them farther away from the by an amount depending upon the speed with As we which they are rotating in their orbits. equal to . B. C.57 times the radius of corpuscles. First suppose the corpuscles are at rest equilibrium . the centre of the triangle coinciding with the centre of the sphere. arranged within the sphere at the corners of an equilateral triangle. when the atom will break up. just balances the radial attraction excited on the corpuscles by the positive the sphere. which will evidently be radial. increase this speed the distance of the corpuscles from the centre of the sphere will increase until at a certain speed the corpuscles will reach the surface of the sphere will cause . further increases in speed them first to rotate outside the sphere and finally leave the sphere altogether. they will be in when they are at such a distance from the centre of the sphere that the repulsion between the corpuscles. In this way we see that the constitution of the ÆTHE RF ORCE .

. for the sake of We of the corbrevity. any would be proportional to each other if the law They very close known as the law of equipartition of energy among the various degrees of freedom of the atom were to apply. ÆTHE RF ORCE . shall. which the mean kinetic energy of the corpuscles inside the atom. no estimate is given of the time required to establish the state con- templated by the law it may be that this time is so long that gases are never able to get into this state. and in the proof given of it in the kinetic theory of gases. in These temperatures are probably not relationship with each other. speak of this kinetic energy as the corpUBfatfar tempuscles within the atom perature of the atom. and the molecular temperature. and we may express the preceding result by saying that the atom will not be stable unless its corpuscular temperature is below a certain value. This law is. to distinguish is We must be careful between cor- puscular temperature. however.QO t7O ELECTRICITY AND MATTER not be permanent. which is the mean kinetic energy due to the motion of the centre of gravity of the atom. if the kinetic aggregate will of the corpuscles inside energy due to the velocity the sphere relative to the centre of the sphere exceeds a certain value. inconsistent with the physical properties of gases.

these positive charges will ultimately be neutralized by ÆTHE RF ORCE . that the centres of gravity of close A B A and B when quite to each other are at rest. under the repulsion separated they will have each a positive charge but as there are now free corpuscles with negative charges moving about in the region in which A and B are situated. etic the increase in the kin- energy is considerable one or more corpuscles. see can easily. rily so.CONSTITUTION OF THE ATOM 99 Let us now . of course. When . A and B will diminincrease. as to make unstable when apart. will A and B unite to form a would do if more complex aggregate as they the corpuscles in them were at rest ? We I think. and they will tend to separate of these charges. A and whose corpuscular temperatures and though not so high. under their mutual attractions. in order to give them the best possible chance of combining. take the case of two aggregations. and suppose. puscle will leave and B may each The departure of a A lose cor- A and B positively charged. the potential en- ergy due to the separation of ish and their kinetic energy will This in- crease in the kinetic energy of the corpuscles in A and B will increase the tendency of and if the corpuscles to leave their atoms. that they will not necessa- For as A and B approach each other. are high.

if they were above a cer- tain limit. and the aggrewould not be formed. If the corpuscu- and before union were temperatures of the corpuscular temperature after union very high.100 ELECTRICITY AND MATTER corpuscles striking against A and B and remain- ing in combination with them. A B would be high also. ÆTHE RF ORCE . point. and incapable of a permanent existence. one congate AB dition for the formation of complex aggregate's is that the corpuscular temperature of their constituents before combination should be sufficiently low. Thus. comIf A B bination may be prevented by the high relative spite velocity of A and B carrying them apart in The of their mutual attraction. the union cannot be permanent. however. the molecula/t* temperature of the gas in which and are molecules is very high. thus conclude that unless the corpuscular temperature after union is less than a certain limit- We ing value. the complex formed being unstable. the corpuscular temperature after union would be too high for stability. the corpuscular and temperature of the aggregate formed by will depend upon the corpuscular temperatures of A B A and B before union. and also upon the diminusystem occa- tion in the potential energy of the sioned lar by the union of A and B. Now.

e. is that even at this temperature the kinetic energy of the corpuscles inside the atom. We shall now proceed these results on to discuss the bearing of the theory that the different chemical elements have been gradually evolved by the aggregation of primordial units. ture. ÆTHE RF ORCE . to take a ment. specific example.e.e. that there no very inti- mate connection between the corpuscular and molecular temperatures. i. that we cannot se- cure the union merely by lowering the molecular . but sure union by lowering it. by cooling the gas union will be impossible unless the corpuscular temperature. is reduced below a cer- We may prevent union by raising the we cannot en- molecular temperature of a gas... i. on the atoms of hydrogen present on why the earth do not combine to form some other ele- Thus.CONSTITUTION OF THE ATOM JQ1 which I wish to emphasize is. temperature. is too great. the reason.. i. even at the exceedingly low temperature at which hydrogen becomes liquid. this view. the kinetic energy due to the motion of the corpuscles inside the atom. and that we may reduce the latter almost to the absolute zero without greatly affecting the former. tain value. It the corpuscular temperamay be useful to repeat here is what we stated before.

The systems of the first kind will combine. This would be so. but we shall proceed We shall anticipate the result of the discussion by saying that it leads to the result that the rate of decay in the corpuscular temperature probably varies greatly from one binary system to another. because when the two units have come together there must be an amount potential energy consequent of kinetic energy produced equal to the diminution in the upon the coalescence of the two units. and thus we shall ÆTHE RF ORCE . as we shall call them. immediately to discuss the way in which this reduction is effected. the corpuscles in the system would have a considerable amount of kinetic energy.102 ELECTRICITY AND MATTER Let us suppose that the first stage has been reached and that we have a number of systems formed by the union of two units. energy of the corpuscles must get reduced. As these binary systems have or- iginally high corpuscular temperatures they will not be likely to combine with each other or with another unit before they can do so the kinetic . When first these binary systems. were formed. Some of the systems will therefore probably have reached a condition in which they are able to combine with each other or with a single unit long before others are able to do so.

in- Thus. then as the universe gets older elements of higher and higher atomic weight may be expected to appear. three. while at the same time there are the appearance of the more complex systems need not be simultaneous with the disappearance of all the simpler ones. will not involve the annihilation of the elements of lower atomic weight. Their appearance. if we regard the systems containing different numbers of units as corresponding to the different chemical elements. however. many of the binary systems left. and we may have systems two or even containing eight units formed before the more persistent of those containing four. The same principle will apply to the formation of further aggregations three or four units . by the systems containing some of these will be ready to unite before the others. since the heavier elements are up of material furnished by hypothesis built the lighter. The number of atoms of the latter will. some of which contain three. Thus.CONSTITUTION OF THE ATOM 1Q3 have systems formed. of course. diminish. one unit have disappeared. others four units. With the further advance of aggregation the number of different sys- tems present at one and the same time will crease. The whole by ÆTHE RF ORCE .

. the lighter atoms. and unless there is disintegration of the heavier tion. number of corpuscles in the atom is ex- ceedingly large.e. however. If. and thus we may have a very at one and the large number of elements existing same time. all be used up at once. i. the atomic weight of the lightest element surviving will continually increase. way the Corpuscles in tlie Atom Lose or Gain Kinetic Energy If the kinetic energy arising from the motion of the corpuscles relatively to the centre of gravity of the atom could by collisions be transformed into kinetic energy as a whole. due to the motion of the atom it into molecular temperature. all aggregations of less than a thousand units have entered into combination and are no longer Tfie free. the lightest known element and hydrogen the atom of hydrogen contains about a thousand corpuscles. however. since is On this view. that the specific heat of a gas at ÆTHE RF ORCE . would follow from the since the kinetic theory of gases.104 ELECTRICITY AND MATTER of the atoms of the latter would not. there is a continual fall in the cor- puscular temperature of the atoms through radia- elements will disappear in time.

as . We ing its have seen. radiating energy and so losing kinetic energy. We therefore. if we have a single corpuscle describing a circular orbit of radius a with uniform velocity v.CONSTITUTION OF THE ATOM 105 constant pressure would be very nearly equal to the specific heat at constant volume whereas. in no gas is there any approach to equality in these specific heats. If the corpuscles and instead of a single corpuscle we had two corpuscles at opposite V ends of a diameter moving round the same orbit with the same velocity as the sinthe loss of energy per second from the gle corpuscle. a matter of fact. that a mov- electrified particle radiates velocity direction. the loss of energy due to radiation per second is 3 =^-5. Thus. that it is not by collisions that the is kinetic energy of the corpuscles diminished. two would be very much less than from the single ÆTHE RF ORCE . energy whenever changing either in magnitude or The corpuscles in the atom will thus emit electric waves. conclude. however (page is 68). The rate at which energy is lost in this way by the corpuscles varies very greatly with the number of the corpuscles and the way in which they are moving. V or where e is the charge on the velocity of light.

Number of corpuscles. ELECTRICITY AND MATTER and the smaller the velocity of the cordiminution in puscle the greater would be the the loss of energy produced by increasing the number of corpuscles.6 4. describ- ÆTHE RF ORCE . taken as one-tenth that of and in the second as one-hundredth. The table applies to is two cases .6 x ID"4 4. is in The radiation from a single corpuscle each case taken as unity.106 corpuscle.10 5.6xlO~17 Thus. we see that the radiation from each of a group of six corpuscles moving with one-tenth the velocity of light of the radiation is less than one-five-millionth part from a single corpuscle. Radiation from each corpuscle.6 x 10~7 1.7x10-^ 1.6 x 10~2 x l(r3 9. The effect produced by is increasing the number of corpuscles shown in the following table.6X10- 5 5.6 x 10-7 4 5 6 1.7 x 10.6 x 10~13 1. which gives the rate of radiation for each corpuscle for various numbers of corpuscles arranged at equal angular intervals round the circular orbit. _F ~~IQ 1 ~ 1 V TOO 1 2 3 9. in one the veloc- ity of the corpuscles light.

If the corpuscles are displaced from the symis metrical position in which they are situated at equal intervals round a circle whose centre at rest. the reduction in the very much greater. number of corpuscles the variation in the rate at which energy is radiated will vary very rapidly with the way the corpuscles are moving about in the atom. of cor- puscular cooling. Thus. the rate of radiation.e.. the rate of radiation will be very much inIn the case of an atom containing a large creased. would be exceedingly small it would vanish altogether if the corpuscles were so close together that they formed a continuous ring of negative electrification. If the same number of particles were moving about irregularly in the atom. then though the kinetic energy possessed by the corpuscles in the second case might be no greater than in the first. i. when the velocity of the corpuscles is only one- hundredth of that of radiation is light. for example. we see that in the radiation of energy from corpuscles whose velocity is not uniform we ÆTHE RF ORCE . Thus.CONSTITUTION OF THE ATOM 1Q7 ing the same orbit with the same velocity. would be immensely greater. while. if we had a large number of corpuscles following close on one an- other's heels round a circular orbit the radiation .

millions of years . for although the corpuscular temperature when the atom of a new element is formed is likely to be exceedingly high. if the view we have been discussing is correct. and the lowering in that temperature required before the atom can enter again into fresh aggregations yet we very large. I will which I shall have to allude later. but a very strong in favor of some such secular changes presumption in the atom. no direct evidence of any change at all in the atom. and so. taking place We must remember. not say a proof of. indeed.10g ELECTRICITY AND MATTER have a process going on which will gradually cool the corpuscular temperature of the atom. that the corpuscles in any atom are receiving and absorbing radiation from other atoms. too. This will tend to raise the corpuscular temperature of the atom and thus help to lengthen the time required for that ÆTHE RF ORCE . enable the atom to form further aggregations and thus tend to the formation of new chemical ele- ments. many thousands. This cooling process must be an exceedingly slow one. I think. that some of the phenomena of radio-activity to afford. have evidence that some of the elements for must have existed unchanged nay. however. we have.

important to realize how large are the amounts of energy involved in the formation of a complex atom or in any rearrangement of the configuration of the corpuscles inside it. enter It is upon fresh changes long before the others. hence. the work required to sep- arate the atom into its constituent units will be ^ '.CONSTITUTION OF THE ATOM \QQ temperature to fall to the point where fresh aggregations of the atom may be formed. If we have an atom containing corpuscles each with a in electrostatic e measured units. comparable with a being the radius of the Thus. the total charge quantity of negative electricity in the atom is n e n and there is an equal quantity of positive elec- tricity distributed through the sphere of positive electrification. the corpuscles are moving The fact much upon different that the rate of radiation depends so the way about in the atom indicates that the lives of the atoms of any particular element will not be equal some of these atoms will be ready to . the formed atom has been by aggregation of these units v 16 ' a will be of the same order of magni- tude as the kinetic energy imparted to those con- ÆTHE RF ORCE . as the sphere containing the corpuscles.

thus : but if m is the mass of a corpuscle and therefore a ma' electrostatic units e _ now when e is measured in m <L = 3 x 10 17 and = 3. let of these atoms in a gram.4 X 1Q- 10 . If Mis the mass of an atom NM= 1. but the following calculation will show what an enormous of kinetic energy the corpuscles in the amount atom must possess even if they have only retained an exceedingly small fraction of that communi- cated to them. Let us calculate the value of * '- a for all the atoms in a N be the number N^'- gram of the substance . ÆTHE RF ORCE . down to the time they became members of the atom under consideration.HO ELECTRICITY AND MATTER stituents during their whole history. from the time they started as separate units. They will in this period have radi- ated away a large quantity of this energy. then is the value of the energy acquired by these atoms.

in the atoms in each We the so shall return to the subject of the internal changes in the atom when we more discuss some of phenomena of radio-activity.CONSTITUTION OF THE ATOM m (1) and therefore N&*y= = 10.2X10 7 X -.e. the greater will be the amount of energy stored up gram. and take for a the value usually assumed in the kinetic theory 8 of gases for the radius of the atom. see. shall begin with the case where the corpuscles We are at rest.. sufficient to lift amount of energy would be a height considerably exceeding one hundred yards. but before doing closely the it is desirable to consider way the corpuscles arrange themselves in the atom. i. 10" cm. then jy (M-= this 1. from million tons through a We (1) that this energy is proportional to the num- lar ber of corpuscles. The corpuscles are supposed to be in a sphere of uniform positive electrification which produces a radial attractive force on each cor- ÆTHE RF ORCE .02 X 10 19 ergs. too. so that the greater the molecuweight of an element. a Let us take the case of the hydrogen atom for which n 1000.

we can see once that they will be in equilibrium if placed so that A B and the centre of the sphere are in the same OB $ the radius of the straight line and OA = = sphere. be in equilibrium of angle with its centre at A B C as O C.57 times the radius of the sphere. and we might suppose that whatever the number of corpuscles the position of equilibrium would be ÆTHE RF ORCE . If there are three corpuscles. If there are four corpuscles equilibrium if placed at the angular points of a regular tetrahedron with its centre at the centre of the sphere. they will an equilateral triand AB OA= OB = OC = (\y. problem is to arrange the corpuscles in the sphere so that they are in equilibrium under this attraction pulsions. In these cases the corpuscles are all on the surface of a sphere concentric with the sphere of positive electrification. these will be in or . 15. Flo. A B. and their mutual re- If there are only two at corpuscles. 16. FIG.112 ELECTRICITY AND MATTER distance from the centre puscle proportional to its the and of the sphere.

One group containing the smaller number of corpuscles is on the surface of a small body concentric with the sphere . to turn to experiment and see if we can make a model in which the forces producing equilibrium are similar to those we have supposed to be at ÆTHE RF ORCE . Such a distribution would indeed technically be one of equilibrium. arranged on the surfaces of concentric shells . say seven or eight at is the most. but a mathe- matical calculation shows that unless the of corpuscles is number quite small. When the number greater than this limiting number. the remainder are on the surface of a larger concentric still body. With any considerable number of corpuscles the problem of finding the distribution when in equilibrium becomes too complex for calculation and we have . and as we go on stages in increasing the number we pass through which more and more groups are necessary for equilibrium. When the number of corpuscles is further increased there comes a stage when the equilibrium cannot be stable even with two groups.CONSTITUTION OF THE ATOM H3 one of symmetrical distribution over the surface of a sphere. this arrangement unstable and so can of corpuscles is never persist. the corpuscles break up into two groups. and the corpuscles now divide themselves into three groups.

Thus the forces on the poles of the floating magnets will be very similar to those acting on the corpuscle in our hypothetical atom . is provided by a negative pole (if the little mag- nets have their positive poles above the water) suspended some distance above the surface of the water. The magnets are placed so that the positive poles are either all above or all below the surface of the These positive poles. parallel to the surface of the water. the projection of the negative pole on the surface of the water. directed to 0. work Such a model is afforded by a very simple and beautiful experiment first made. ÆTHE RF ORCE . I think. will be radial. The attractive force water.114 ELECTRICITY AND MATTER in the corpuscle. by Professor Mayer. This pole will exert on the positive poles of the little floating magnets an attractive force the component of which. In this experiment a number vessel of water. of little magnets are floated in a The magnets are steel needles floated magnetized to equal strengths and are by being thrust through small disks of cork. and if the negative pole is some distance above the surface the component of the force to O will be very approximately proportional to the distance from O. other with forces varying inversely as each repel the distance between them. like the corpuscles.

CONSTITUTION OF THE ATOM H5 the chief difference being that the corpuscles are free to move about in all directions in space. From this will be seen that when the number of five the floating magnets does not exceed magnets ÆTHE RF ORCE . The configurations which the floating magnets assume as the number of magnets increases from two up to nineteen is shown in Fig. which was given by Mayer. The configuration taken up when the magnets are more numerous can be found from the follow- ing table. FIG. 17. which table it is also due to Mayer. 17. while the poles of the floating magnets are constrained to move in a plane parallel to the surface of the water.



arrange themselves at the corners of a regular polygon, five at the corners of a pentagon, four at
the corners of a square and so on.





greater than

five this

Thus, six longer holds. themselves at the corners of a hexagon, but divide into two systems, one magnet being at the centre

arrangement no magnets do not arrange


five outside it at

the corners of a regular pentalasts until

This arrangement in two groups

there are fifteen magnets,

groups groups and so on.

when we have three with twenty-seven magnets we get four

Arrangement of Magi








but reappear Let us on.
the floating
of magnets

when we reach potassium, and so now consider the arrangements of magnets, and suppose that the number
proportional to the combining weight


of an element.



any property were


ciated with the triangular arrangement of magnets,

would be possessed by the elements whose combining weight was on this scale three, but would
not appear again until

weight ten, when it we have the triangular arrangement in the middle and a ring of seven magnets outside. When the

reached the combining reappears, as for ten magnets



of magnets


increased the


arrangement disappears for a time, but reappears

with twenty magnets, and again with


the triangular arrangement appearing and dis-

appearing in a way analogous to the behavior of the properties of the elements in the Periodic Law. As an example of a property that might

very well be associated with a particular grouping of the corpuscles, let us take the times of vibration of the system, as

shown by the

position of


the lines in the spectrum of the element. First us take the case of three corpuscles by themselves in the positively electrified sphere.


three corpuscles have nine degrees of freedom, so


while corresponding to I> there are two equal roots. Suppose that the lines in the spectrum of the a. one ÆTHE RF ORCE . we suppose that there is only one root giving the period corresponding to the line -4. these periods in this case would be HQ Some of infinitely long. so that we should not get nine dif- ferent periods.. three equal roots corresponding to <7. and several of the possible periods would be equal to each other.e. regarding the periods as given by an equation with nine roots. number of periods which coalesce at that line i. three corpuscles are as represented in Fig. 18.CONSTITUTION OF THE ATOM that there are nine possible periods. where the figures under the lines represent the . 18 A a c D c e 3 / 2 A B c a FIG.

instead of being alone in the sphere. the size of the sphere in which they are placed. corre- ÆTHE RF ORCE . 10. just as the triangle of mag3. The absolute values of the periods would generally be entirely different. These periods would relations to each other. will it does not affect the effect have a great upon the absolute value of any one of them. these quantities. suppose that these three several groups in corpuscles. and 35 magnets. innumerical have certain dependent of the charge on the corpuscle. Now. Let us consider how the the presence of the other groups would affect periods of vibration of the three corpuscles. and two to E. When self there the group of three corpuscles was by itwere several displacements which gave the same period of vibration . or their Each of distance from the centre of the sphere. to ELECTRICITY AND MATTER O. although ratio of the periods.120 root. nets forms a constituent of the grouping of 20. and although it might be modified it would not be destroyed. for example. Using the phraseology of the Planetary Theory. but the relationship existing be- tween the various periods would be much more persistent. we may regard the motion of the three corpuscles as " disturbed " by the other groups. form but one out of it.

the distances between the components of the doublets and triplets increasing with the atomic weight of the elements. Thus. the element which contain this particular grouping of corpuscles as being in the same group in the classification of elements according to the Periodic Law. instead of being a single line. in the spectrum. there are other different displacements will groups present. They would. 18 Thus. come 18 doublets. The investigations of Rydberg. the spectrum would now resemble Fig. b'. would become a triplet. all C there giving the same period. is shown in Fig. were three displaceWhen. however. Thus. we should get in the spectra of these elements homologous series of lines. Runge and Pas- ÆTHE RF ORCE . if we regarded . how^ ever.CONSTITUTION OF THE ATOM 121 spending to the line ments. <7. while B and E would beA D would remain single lines. then these no longer be sym- metrical with respect to these groups. be very nearly equal unless the effect of the other groups is very large. so that the three periods will no longer be quite equal. The appearance as the number of groups increases c. the more groups there are surrounding the group of three the more will the motion of the latter be disturbed and the greater the separation of the constituents of the triplets and doublets.

so that the passage from the that of configuration of greater potential energy to to the kinetic less would give corpuscle. of the values of the potential energy stored which we gave an estimate on page 111. we infer that a change by even a small would develop an amount of kinetic energy which if converted into heat would greatly transcend the amount of fraction in that potential energy heat developed when the atoms undergo any chemical combination.122 ELECTRICITY AXD MATTER chen and Keyser have shown the existence in the series of spectra of elements of the same group having properties in many respects analogous to those we have described. five form two groups. fourteen magnets . these configurations correspond to different amounts configuration of potential energy. energy From in the atom. ÆTHE RF ORCE . . magnets form two. lines Another point of periments is interest given is that there for the by Mayer's exmore than one stable same number of magnets. twenty-eight four. fifteen three twenty seven magnets form three groups. known An inspection of the table shows that there are with the number certain places in it where the nature of the con- figuration changes very rapidly of magnets while six magnets form one group. thus.

without. in many cases. If we arrange the chemical elements in the order of their atomic weights we find there in proper. however. they will. we and rubidium and so This effect seems analogous to that due to the regrouping of the magnets. be situated at the corners of an equi- angular triangle . Then there until is more or less continuity in the properties get to chlorine. as when they are at rest. which is followed by potassium. be rotating round the centre of the sphere. and the distance of the corpuscles from the centre will be ÆTHE RF ORCE . for example. Thus. for example. if we have three corpuscles in the sphere. they are in a state of steady motion and describing circular orbits round the at rest if. the effect of the centrifugal force arising from this motion will be to drive the centre of the corpuscles farther away from the sphere. centre of the sphere. the next break occurs at bromine on. destroying the character of the configuration. in the state of steady motion. this triangle will. are certain places ties of consecutive where the difference elements is exceptionally great thus.CONSTITUTION OF THE ATOM 123 and so on. we have supposed the corpuscles to be however. we have extreme fluorine differences in properties between and sodium. So far .

The arrangement and there will become unin the sys- be a convulsion ÆTHE RF ORCE . if These. stable. rotating rapidly. There are. and the corpuscles tend to place themselves at the corners of a regular tetrahedron. take the case of four corpuscles. however. the plane of the square being at right angles to the axis of when. many cases in which rota- tion is essential for the stability of the configuration. which is the stable arrangement when the cor- puscles are at rest. Thus. the top like the corpuscles being its velocity of rotation exceeds critical a certain value. the square arrangement will persist until the velocity of the corpuscles is reduced to the will then critical value. are in stable steady motion when at the corners of a square. The system of four corpuscles at the corners of a square may be compared with unstable unless a spinning top. however.124 ELECTRICITY AND MATTER at rest and will ingreater than when they are crease with the velocity of the corpuscles. the velocity of rotation below a certain value. but that in some way or another the corpuscles gradually lose their kinetic energy. the arrangement of four corpuscles in one plane berotation . Let us suppose that corpuscles exceeds initially the velocity of the this value. of the corpuscles falls comes unstable.

In such cases the configuration when the corpuscles are rotating with great rapidity will (as in the case of the four corpuscles) be essentially different from the configuration of the rest. with special atomic weights. below the is critical value. such that.CONSTITUTION OF THE ATOM tern 125 accompanied by a great evolution of kinetic energy. These considerations have a very direct bearing on the view of the constitution of the atoms which we have taken in this chapter. a configuration is stable. same number of corpuscles when at Hence there must be some critical velocity critical of the corpuscles.e.. which becomes unstable when the velocity is reduced below the critical value. for they show that with atoms of a special kind. i. the corpuscular cooling caused by ÆTHE RF ORCE . for velocities greater than the one. instability and there a kind of convulsion or ex- accompanied by a great diminution in the potential energy and a corresponding increase in the kinetic energy of the corpuscles. When the velocity sinks sets in. in the kinetic energy of the corpuscles sufficient to may be detach considerable numbers of them from the original assemblage. This increase plosion. Similar considerations will apply to many assemblages of corpuscles.

On formed the view that first the lighter elements are aggregation of the unit the of which is the corelement doublet. negative and that it is the combination of the puscle. we should expect the corpuscles in the heavy atoms to be ar- ranged as were in bundles. each bundle corresponding to ÆTHE RF ORCE . due to the arrangement of the corpuscles atom. by by the atoms of the lighter elements that the atoms of the heavier elements are produced. this It would cause the atom to emit energy from derived the potential energy energy being . produce instabilthe atom. in the We shall see when we consider the pheis nomenon of radio-activity that there a class of to those bodies which show phenomena analogous just described. and produce such an in- crease in the kinetic energy of the corpuscles as to give rise to greatly increased radiation. might.126 ELECTRICITY AND MATTER the radiation from the moving corpuscles which we have supposed to be slowly going on. when it ity inside reached a certain stage. the arrangement of the corpuscles in each bundle being similar to the it arrangement in the atom of some lighter element. In the heavier atom these bundles would act as subsidiary units. and it might be detachment of a portion of the atom.

in this model of an atom any scope for the electro-chemical properties of the real atom such properties. dyads. called monads. the atom to ? suggest the possession of such a property as that called by the chemists valency i.. for example.CONSTITUTION OF THE ATOM J27 one of the magnets in the model formed by the floating magnets. as those . while inside the bundle themselves the corpuscle would be the analogue to see of the magnet. the atom of chlorine a negative charge of electricity is there anything in the model of Again. in while in a compound formed by an element the molein the second. the property which enables us to divide the elements into . triads.e. such that in a compound formed by any two elements of the first contain the same group the molecule of the compound will number of atoms of each element. for example. electro-positive and is electro- Why. illustrated by the division of the chemical ele- ments into two negative. the first group with one A B ÆTHE RF ORCE . classes. groups. We built sess must now go on in the whether an atom could posIs up any way we have supposed of the properties of the real atom. does an atom of sodium or potassium tend to acquire a positive. if this the con- stitution of the atom. there. for example.

12 g cule of the ELECTRICITY AND MATTER contains twice as ? compound many atoms of Let us atom. A as of B. collision They may be detached also by atom with other rapidly moving atoms or free corpuscles. When once a corpuscle has escaped from an atom the latter will have a posThis will make it more difficult for itive charge. may even be that after one has escaped. can readily conceive that the ease with which a par- escape from. and so on now turn to the properties of the model It contains a very large number of corpus- cles in rapid motion. a second negatively electrified corpuscle to escape. These may escape ity owing to to their high veloc- enabling them of the travel beyond the attrac- tion of the atom. the attraction of the ÆTHE RF ORCE . We have evidence from the phenomena electricity connected with the conduction of through gases that one or more of these corpuscles can be detached from the atom. or be knocked out an atom may vary very much in the atoms of the different elements. Now we of. In some atoms the velocities of the corpuscles may be so great that a corpuscle It escapes at once from the atom. for in consequence of the positive charge on the atom the latter will attract the second corpuscle it more strongly than ticle will did the first.

but three were. would by aggregation with these corpuscles ceive a negative charge. The magnitude of the negative charge would depend upon the firmness with which the atom held its corpuscles.CONSTITUTION OF THE ATOM positive electrification thus left on the 129 atom is not sufficient to restrain a second. Thus. Atoms of this kind if placed in a region where corpuscles were present re. negative charge on the atom would be one If two corpuscles were not sufficient to expel a corpuscle. the maxi- mum unit. or three corpusOn the other hand. nay. and so on. according as they lost one. or three units. Such atoms would ac- quire positive charges of one. there may be atoms in which the that few. two. the atoms of this class tend to get charged with ÆTHE RF ORCE . If a negative charge of one corpuscle were not sufficient to expel a corpuscle while the negative charge of two corpuscles could do so. the maximum negative charge would be two units. corpuscle from escaping. two. cles. corpuscles escape of their own accord. or even a third. velocities of the corpuscles are so small if any. they may even be able to receive one or even more than one corpuscle before the repulsion exerted by the negative electrification on these foreign corpuscles forces any of the original corpuscles out.



negative electricity and correspond to the electronegative chemical elements, while the atoms of the



considered, and which readily lose

corpuscles, acquire

a positive charge and corre-

spond to the atoms of the electro-positive elements.


might conceive atoms in which the equilibrium of the corpuscles was so nicely balanced
that though they do not of themselves lose a corpuscle,

and so do not acquire a positive charge, the

repulsion exerted

by a

foreign corpuscle coming
sufficient to drive out a

on to the atom would be

Such an atom would be incapable of

receiving a charge either of positive or negative

Suppose we have a number of the atoms that readily lose their corpuscles mixed with a number of those that can retain a foreign corpuscle. Let us call an atom of the first class A, one of the second

and suppose that the


atoms are of

the kind that lose one corpuscle while the IB atoms
are of the kind that can retain one, but not


than one


then the corpuscles which escape from
there are an equal

A atoms will ultimately find a home on the B

atoms, and


of the

two kinds
ly all the

of atoms present


shall get ultimate-


atoms with the unit positive charge,






atoms with the unit negative charge.

These oppositely electrified atoms will attract each other, and we shall get the compound
that lost
If the



atoms had been of the kind



and the


atoms the


as before, then the


charge of two positive units, of one unit of negative electricity.
a neutral system

atoms would get the the atoms a charge



of the


Thus, to form atoms must com-

bine with one of the A's and thus the



HI would be formed.
Thus, from this point of view a univalent elecatom is one which, under the circum-


prevailing when combination is taking has to lose one and only one corpuscle beplace, fore stability is attained a univalent electro-negstances




one which can receive one but not

more than one corpuscle without driving off other corpuscles from the atom; a divalent electroloses two corpuscles and positive atom is one that
no more, and so on. The valency of the atom thus depends upon the ease with which corpuscles can escape from or be received by the atom



may when

be influenced




taking place.


would be

easier for a corpuscle,

when once




had got outside the atom, to escape being pulled back again into it by the attraction of its positive electrification, if the atom were surrounded by good
were isolated in space. We can understand, then, why the valency of an atom may in some degree be influenced by the physical
conductors than
if it

conditions under which combination


taking place.


the view that the attraction between the

atoms in a chemical compound is electrical in its to enter into origin, the ability of an element
chemical combination depends upon its atom having the power of acquiring a charge of electricity.
This, on the preceding view, implies either that the

uncharged atom


unstable and has to lose one or

more corpuscles before
state, or else that
it is

so stable that

can get into a steady it can retain

one or more additional corpuscles without any of the original corpuscles being driven out. If the
range of stability

such that the atom, though

when uncharged, becomes



receives an additional corpuscle, the


will not

be able to receive a charge either of positive or negative electricity, and will therefore not be able
to enter into chemical combination.

Such an atom

would have the properties of the atoms of such
elements as argon or helium.


a trivalent at the end of and so on is and that when the chemical compound represented by a graphic formula in this way. The theory of bonds when represented graphically supposes is. identical that from each (the symbol is of a bond) univalent atom a straight line proceeds. however. view of Davy and of declared that Helmholtz. . a univalent atom has one unit charge. each atom must be at the end of the proper number of the lines which represent the bonds. when regarded in one aspect almost with the electrical theory. the mightiest of the chemical forces are electrical in their origin. Now. if we take as our unit of the atom is charge the charge on the corpuscle . on the electrical view of chemical combination. was first proposed by Berzelius Faraday. to have little use of this idea. . it was also the too. having found the conception of "bonds of apparently This doctrine of bonds affinity" more fruitful. Chemists in general seem. therefore the beginning or end of one unit Faratube the beginning if the charge on the day : ÆTHE RF ORCE . a divalent lines.CONSTITUTION OF THE ATOM 133 The view that the forces which bind together the atoms in the molecules of chemical compounds are electrical in their origin. atom atom at the end of two such three. made but however.

is On the electrical theory. positive. if we interpret the "bond" of the chemist as indicating a unit Fara- day tube. the structural formulae of the chemist can electrical theory. there a difference between the ends. the symbol indiis not re- garded as having direction no difference is made on this theory between one end of a bond and the other. however. the other to a negative An example or two may perhaps be the charge. connecting charged atoms be at once translated into the in the mole- cule. Thus. as one end cor- responds to a positive. Let us take the gas ethane whose structuis ral formula written According to the chemical view there is no differ- ÆTHE RF ORCE .134 ELECTRICITY AND MATTEB is atom tive. the end if the charge is nega- A divalent atom has two units of charge and it is therefore the origin or termination of two unit Faraday tubes. easiest way of indicating the effect of this consid- eration. one point of difference which little deserves a consideration : cating a bond on the chemical theory . There is. however.

one of the carbon atoms will have a charge of four positive units. i.. there is bon atom to the positive one tube which goes from one carThis means an additional other.e. however.e. there would. while the other will have a charge of three positive and one negative unit. Thus.. be a difference on the electrical view. But in addition to the Faraday tubes coming from the hydrogen atoms. that on this view the two carbon atoms are not in the same state. exist A still greater difference must is between the atoms when we have what i. two positive units so . For all let us suppose that the hydrogen atoms are negatively electrified. charge on one carbon atom and a negative charge on the other. the compound ÆTHE RF ORCE . called double linking. the three Faraday tubes going from the hydrogen atoms to each carbon atom give a positive charge of three units on each carbon atom. when the carbon atoms as in are supposed to be connected by two bonds.CONSTITUTION OF THE ATOM 135 ence between the two carbon atoms in this com- pound .

and in this case chemists find neces- ÆTHE RF ORCE . 6' 2. properties which can be the chemical constitution of the molecule let known.e. then p is the value of . well-known instance of the additive proppower of different substances it erty is the refractive for light. A B C represent the atoms of if three chemical elements. I some physical constant for the molecule of q the value for 13%. Thus.136 ELECTRICITY AND MATT K if II one carbon atom had a charge of four positive units. its value for a mole- whose chemical composition represented by the formula A & J2y Cz is \px-\r\qy-\-\rz. which are known as additive prop- when is calculated i. positive and two negative units. then if this con- stant obeys the additive law.. and r for cule of the substance is . the other would have a charge of two Here. If the atom in different to different values of compounds x y z are to use A occurs in different states compounds we should have different values of A p for these compounds. We can only expect relations like this to hold when the atoms which occur in the different corresponding the same. We might expect to discover such differences as are indicated by these considerations by the in- vestigation of erties.

linked with another atom as when. On the we should expect that. It it is not linked with another carbon may be urged that although we can conceive that one atom in a compound should be positively and the other negatively electrified when the atoms are of different kinds. ceteris paribus. an atom in a gas rapidly moving atoms or which keep striking against it may corpuscles have corpuscles driven out of it by these collisions when surrounded by and thus become positively other hand. depending as it does on the power of the atom to emit or retain corpuscles. atom at all. They use. electrified. for the refraction of the carbon atom when singly is a carbon atom according as the atom singly linked. With reference we may remark that the electrical state of an atom. for an example. may be very largely influenced the atom. it is not easy to do so when the atoms are of the same kind. point OK JVj and so on. the atom would be less likely to lose a corpuscle when it is in a gas than when in a solid or a ÆTHE RF ORCE .CONSTITUTION OP THE ATOM sary to use different values for the refraction 137 due doubly or the same value however. as in the com- pound G HI. by circumstances external to Thus. as they the elementary gases to this are in the molecules of HI.

however. ELECTRICITY AND MATTER For when the atom in a gas after a corpuscle has it just left velocity to rely has nothing beyond its own from the attraction upon to escape of the positively electrified atom. atoms are too far away to exert any forces upon When. These charged corpuscles when acted upon by an electric force are set in motion rent. mercury is in the gaseous conductivity has been shown by by the same number of Strutt to be an exceedingly small fraction of the conductivity possessed molecules when gaseous. since the other it. help it has left to avoid falling back again into atom. state. however. effect As an instance of this we may is liquid and gaseous take the case of mercury in the In the liquid state states. the atom is in a liquid or a the attractions of the other atoms which this its solid. crowd round cle atom may. the and constitute an electric cur- conductivity of the liquid mercury inthe dicating presence of a large number of cor- puscles. its electrical When. when once a corpusatom. One is way of regarding this electrical conductivity to suppose that corpuscles leave the atoms of the mercury and wander about through the interstices between the atoms. mercury a good conductor of electricity.138 liquid. We have thus indications ÆTHE RF ORCE .

others negatively electrified. This argu- ment would not apply to very electro-positive we should not expect to form molegases. possess considerable ÆTHE RF ORCE . if the atoms were not too electro-positive to be able to retain a negative charge even when in the gase- ous state. but since there would puscles in the gas we should expect them to electrical conductivity. Thus. tend to find a home on the more slowly moving atoms. the more rapidly moving ones. likely to lose corpuscles than the slower The faster ones would thus by the positively loss of their corpuscles become electrified. those with changes of opposite signs would combine to form a diatomic molecule. Suppose then that we had a great number of atoms all of one kind in the gaseous state and thus mov- ing about and coming into collision with each other. since they would make the most violent collisions. would be more ones. These be many free corcules. while the corpuscles driven off would.CONSTITUTION OF THE ATOM that the 139 atoms even of an electro-positive sub- stance like mercury may only lose comparatively few corpuscles when in the gaseous state. some of the atoms would and get positively.

and make a gas through which they pass a conductor of electricity. like Rontgen and cathode rays. were more radio-active than equal volumes of pure uranium. and M. and that Mme. This in- dicated that these minerals contained a substance or substances much more radio-active than uran- ium itself. In 1898 Schmidt discovered that thorium possesses similar This power of emitting rays is called properties. and a systematic attempt was made to ÆTHE RF ORCE . radio-activity.CHAPTER VI RADIO-ACTIVITY AND RADIO-ACTIVE SUB- STANCES IN 1896 Becquerel discovered that uranium and its salts possess the power of giving out rays which. Curie found these. affect a photographic plate. This property of uranium led to a careful examination of a large number of minerals containing this substance. although only a fraction some of of these minerals consisted of uranium. and substances which possess the power are said to be radio-active. and notably some specimens of pitch-blende.

M.RADIO-ACTIVE SUBSTANCES isolate these substances. that a similar gas is contained in the soil. the bismuth. They deter- succeeded in isolating the first of these and mined 225. with the collaboration of MM. combining weight. dying away in some months after its preparation. and closely resembling it in its chemical properties polonium associated with . and with the changes that go on in ÆTHE RF ORCE . wells contain a specimens of water from deep have found Geitel and Elster and radio-active gas. Curie. Neither polonium nor actinium has yet been isolated. skill conducted with marvellous and perseverance. which was found to be Its spectrum has been discovered and examits ined by Demarcay. nor have their spectra The activity of polonium has been observed. succeeded in establishing the existence of three new radio-active substances in pitch-blende : radium associated with the ba- rium in the mineral. Bemont and Debierne. These radio-active substances are not confined I have lately found that many to rare minerals. 141 After a long investigation. been found to be fugitive. and actinium with the thorium. These radio-active substances may be expected to be of the greatest possible assistance in the task of investigating problems dealing with the nature of the atom. and Mme.

made up a. in the proportion of a second to thousands of years. Character of the Radiation Rutherford found that the radiation from uran ium. The quantity of these substances which can be detected to the corresponding amount of the other elements which have to be detected by the ordinary is methods of chemical analysis. of three distinct types which he calls the and y radiations. possessed by these substances are so For the properties marked as to make the detection of exceedingly minute quanti- ties of them a matter of comparative ease. The a radiation is very easily absorbed. could with radio-active substances prove apprecifore able effects in the course of a few hours. changes which would have to go on for almost geological epochs with the non-radio-active substances. /3. while the all. and it has subsequently been found that the is same is true for thorium and radium. Thus. bethey became large enough to be detected. the /8 radiation is much more penetrating. y radiation is the most penetrating of Investigations of the ÆTHE RF ORCE .142 ELECTRICITY AND MATTER the atom from time to time. being unable to penetrate more than a few millimetres of air at atmospheric pressure.

the value of m . that the ratio of m is is 6 X 10 3 and the velocity of these particles . and the direction of deflection He finds. the direction of the deflection show- ing that the rays carried a charge of negative tricity. He determined. but Rutherford has recently shown that they can be deflected. the velocity for some of the rays was more than two. positive charge. shows that they carry a. using the method described in Chapter IV. and his measurements have been con firmed by Des Coudres. .RADIO-ACTIVE SUBSTANCES effects of 143 magnetic and electric forces on these three types of radiation have of entirely that the ft shown that they are different characters. Becquerel showed rays were deflected by electric and magelec- netic forces. the ratio of the charge to the mass of the carriers of the negative T electricity he found that it was about 10 and that .third s that of light. He thus proved that the rays consisted of corpuscles travelling at prodigious speeds. ft The a the ft rays are not nearly so easily deflected as rays. 2 X 10* have centimetres per second. The value of shows that the carriers of the positive electrification ÆTHE RF ORCE .

masses comparable with those of ordinary atoms thus m for hydrogen is 104 and for helium 2. The very high velocity with which these are shot out involves an enormous expenditure of ento ergy.!44 ELECTRICITY AND MATTER . indicating either that radium is a compound containing lighter elements or else that the atom of radium is disintegrating into such elements. There considerable resemblance between a radio-active substance and a substance emitting secondary radiation under the influence of Rontgen rays : the secondary radiation ($ is known . a point which we shall return later. The y rays. The value of m for the a rays obtained by Rutherford and Des Coudres suggests the existence of a gas heavier than hydrogen but lighter than helium. as far as we know. it is possible ÆTHE RF ORCE .5 X 10 3 . to contain radiation of the part of the radiation is and y types and as exceedingly easily absorbed. being unable to penetrate more than a millimetre or so of air at atmospheric pressure. is are not deflected either by mag- netic or electric forces. One is of the most interesting things about this result that the value of m shows that the atoms shot off are not the atoms of radium.

Rutherford called it the " emana- The emanation can pass through water or the strongest acid and can be raised to tempera- ÆTHE RF ORCE . Emanation from Radio-Active Substances Rutherford proved that thorium emits something which is radio-active and which is wafted about by currents of air as if it were a gas . i. particles. liberate the energy locked up in the atom.j are present also. This analogy raises the question as to whether there may not. for might lead to a way of doing by external agency what radio-active bodies can do spontaneit ously. energy being derived from changes taking place in the atoms of the body exposed to the Rontgen rays. This point seems worthy of investigation." exists. in order to avoid prejudging the question as to the physical state in which the substance given off by radium tion.e.RADIO-ACTIVE SUBSTANCES that closer investigation positively electrified 145 may show that a rays. in the case of the rays. body struck by energy case of the the Rbntgen be a liberation of such as we shall see occurs in the radio-active substances.e. i.. the energy emitted by the radiating substances being greater than the energy in the Rontgen rays falling upon it this excess of .

by the kindness of Professor Dewar. cient of diffusion of the air radium emanation through has been measured by Rutherford and Miss Brooks and they concluded that the density of the emanation was about eighty. taking about four days to sink to half its activity. the latter which is almost always found associated with thorium. The emanation of radium has been liquefied by Rutherford and Soddy . . been able to liquefy the radio-active gas found in water from deep wells. They diffuse gradually through air and other gases. There seems every reason for thinking that those emanations are radio-active matter in the gaseous form they can be wafted from one place to another by currents of air like a gas they diffuse through a porous plug at a rate which shows . and I have. sinking to half its value in about one minute. The radio-activity of the thorium emana- tion is very transient.146 ELECTRICITY AND MATTER tures at which platinum is incandescent without suffering any loss of radio-activity. In this inertness it of resembles the gases argon and helium. which very ÆTHE RF ORCE . The coeffiis that their density very high. The a Curies found that radium also gives off radio-active emanation which is much more persistent than that given off by thorium.

It is true that they are not capable of detection by any chemical tests of the ordinary type. but gets entangled in the salt and accumulates. Each portion of a salt of radium or thorium is giving off the emanation. nor can they be detected by spectrum analysis. quite possibly identical with tions In short the emana- satisfy every test of the gaseous can be applied to them. It is not. but this is only seem to state that because they are present in very minute quantities quantities far too small to be detected even by spectrum analysis. aggeration to say that certainty it is possible to detect with by the electrical method a quantity of a radio-active substance less than one-hundred-thou- sandth part of the least quantity which could be detected by spectrum analysis. a method of detection which is exceedingly rough when compared with the electrical methods which we are able to employ for an ex- radio-active substances. whether that portion be on the inside or the outside of the salt. does not escape into the air. I think. is If such a radiois active salt dissolved in water. however. the emanation coming from the interior of a salt.RADIO-ACTIVE SUBSTANCES closely resembles the 147 is emanation and it. there at first a has been great evolution of the emanation which ÆTHE RF ORCE .

and this induced activity can be detected on negatively electrified ÆTHE RF ORCE . Induced Radio-Activity Kutherford discovered that substances exposed from thorium become radio-active.148 ELECTRICITY AND MATTER stored up in the solid salt. Thus. The induced radio-activity is especially de- veloped on substances which are negatively electrified. becomes as radio-active as metal when placed in contact with the emanations of thorium or radium. This phenomenon called in- duced radio-activity. The amount of induced radio-activity does not depend upon the nature of the substance on which it is induced thus. the induced radio-activity concentrated on the negatively electrified wire. paper . in is which a negatively electrified wire is placed. The emanation can be by boiling the extracted from the water either water or bubbling air through it. the emanais to the emanation and the Curies discovered almost simultaneously that the same property tion is possessed by from radium. The stored up emanation can also be driven off from salts in the solid state by raising them to a very high tem- perature. if the emanation is contained in a closed vessel.

eleven hours. one fall.RADIO-ACTIVE SUBSTANCES bodies ^49 when it is too electrified surfaces. this is afforded Further evidence of by an exin- periment made by Miss Gates. weak to be detected on unThe fact that the nature of the induced radio-activity does not depend on the substance in which it is induced points to its being due to a radio-active substance which comes is deposited it from the emanation on substances with which in contact. as in the case much less emanation due to actinium is said only to be active ÆTHE RF ORCE . taking about four days instead of one minute to fall to half its value. in fine which the duced it wire was. for from that due to the radium emana- whereas the activity of the thorium emaso transient drops to half its value in one minute. in about forty minutes instead ing to half its value The of thorium. driven off the wire and de- radio-activity on a posited on the surrounding surfaces. gives rise to a very durable induced radio-activity. by raising to incandescence. of. radio-activity The induced is due to the thorium emanation very different tion. the induced radio-activity due to it takes about eleven hours to fall in the nation is that it same proportion. The emanation due to radium. which is much more lasting than the thorium emanation.

When this separation has been effected the thorium left be- hind is for a time deprived of most of ity. the the 7i T X has is thorium X. The it radio-activity of the thorium X slowly decays while While this has that of the rest of the thorium increases until has recovered its original activity. but the induced it radio-activity due to seems to be nearly as permanent as that due to radium. in a most interesting and important investigation.150 for a ELECTRICITY AND MATTER few seconds. the radio-activity of the Th JThas vanished. its radio-activ- which is now to be found in the T h X. view that the radio-active part of the thorium. Separation of the Active Constituent from Thorium Rutherford and Soddy. continually being produced from the ÆTHE RF ORCE . which they showed could be separated from the rest of the thorium by chemical means. The time taken for the radio-activity of the T h X to die away to half its original value has been shown by Rutherford and Soddy to be equal to the time taken by the thorium from which been separated to recover half its All these results support the original activity. been going on. have shown that the radio-activity of thorium is the thorium into a form which they call due to the passage of T h X.

it. ity of the thorium X. we have only the ordinary methods of chemical analysis to rely upon. You will have noticed how closely. might yield important able that helium is almost invariably a It is remark- constitu- ent of these minerals. steadily dies away. as we may accumulating but inasmuch as it has the time in the thorium. so that if the activity of thorium X were permanent. the radio-activity of the tho- rium would continually increase. lost its radio-activity. The radio-activ. and as these are almost infinitely less delicate than the tests it we can apply to radio- active substances. as pointed ÆTHE RF ORCE . The question arises as to what becomes of the Tli X and the emanation when they have This dead all lost their call radio-activity. is ThX. however. which will reach a steady the increase in the radio-activity due h is balanced by to the production of fresh value when T X the decay in the activity of that already produced. seems possible that a careful examination of the minerals in which thorium and radium occur It information. This prevents the unlimited increase of the radioactivity of the mixture. might take almost geological of the dead TJiX epochs to accumulate enough to make detection possible by chemical analysis.RADIO-ACTIVE SUBSTANCES 151 thorium itself .

which is the one on which we have the fullest information.we have first the change of thorium into thorium X. On this view the substance while radio-active is continually being transformed from one state to another. The very large amount of is strik- energy emitted by radio-active substances ingly shown by some recent experiments of the Curies on the salts of radium.152 ELECTRICITY AND MATTER out by Rutherford and Soddy. salts give They find that those out so much energy that the absorption of this by the salt itself is sufficent to temperature of the salt the air by a very appreciable amount their experiments as keep the permanently above that of in one of It much as 1.5 C. of the emanation is activity accompanied by a further transformation. then the change of the thorium X into the emanation and The radiothe substance forming the a rays. These transformations may be sufficient ac- companied by the liberation of to supply that carried off energy emits by the rays it while radio-active. the production of radio-activity seems connected with changes tak- Thus. one of the products being the substance which produces induced radio-activity. take the case of thorium. to ing place in the radio-active substance. appears from their measurements that a gram of radium ÆTHE RF ORCE .

the views we have just explained are true. superior limit of the let us make the assumption that the whole mass of radium gets transformed into the a parthe emanaticles (as a matter of fact we know that ÆTHE RF ORCE . this stock is continually being replenished by the transformation of other chemical elements into radium. If.RADIO-ACTIVE SUBSTANCES 153 gives out enough energy per hour to raise the temperature of its own weight of water from the freezing to the boiling point. indeed. however. a rough guess as to the probable duration of a sample of radium by combining the result We may make that a gram of radium gives out 100 calories per hour with Rutherford's result that the a rays are particles having masses comparable with the mass of an atom of hydrogen projected 10 9 centimetres per with a velocity of about 2 X second by for let us suppose that the heat measured . and the stock of radium its is evolution must cease when exhausted . unless. and to get a to the time the radium will last. this energy arises from the transformation of radium into other forms of matter. the Curies is due to the bombardment of the salt radium by these particles. This evolution of energy goes on uninterruptedly and apparently without diminution.

radium to be of the order of 50. the energy developed in these transformations must be on a very much greater scale than that developed in any known chemical ÆTHE RF ORCE .2 X 9 .2 X the radium during If its life is x X 4. From this estimate we should expect the life of a piece of years. Nm = X 10'. 10 9 ergs. or about 50.15 4 ELECTRICITY AND MATTER is tion a? produced as well as the a life in particles). a result which shows that if this is energy derived from transformations in the state of the radium. jVis the number of a particles emitted in this the mass of one of them in grams. 10* but if the gram of radium 1. and by Rutherford's experiments v = 2 hence we have = 4 | Jg V * 10 = 18 ^ g 10' hours.2 X 10 is X 4.2 X 10 9 ergs. then the energy in the a particles is J Nmv*. In the course of its life the 5 gram X of radium will have given out about 10 10 calories. converted into the a particles.000 This result shows that we could not expect to detect any measurable changes in the space of a few months.000 years. m the velocity. Let . be the hours of a grain of radium then since the gram emits per hour 100 calories. the amount of energy emitted by or 4. but ergs. this is to be equal to a? hence Nm v* = x X 4. v time.

which arrangement becomes unstable when the energy sinks below a certain value and is succeeded by another configuration. while in the are atomic. 155 On the view we have taken the difference between the case of radium and that of ordinary chemical reactions is that in the latter the case of ra- changes are molecular. gradually to decrease. I think. critical value. then. The dium the changes example given on page (111) shows how large an amount of energy may be stored up in the atom if we re- gard it as built up of a number of corpuscles. being of the nature of a decomposition of the elements. when it energy were reached the the top would become unstable and down. We may. velocity are stable when arranged in a certain way. get some light on the processes going on in radium by considering the behavior of a model atom of the kind described on page and which may be typified by the case of the corpuscles which when rotating with a high 124. top spinaxis is another model of the about a vertical ning A same type.RADIO-ACTIVE SUBSTANCES reactions. would fall ÆTHE RF ORCE . position if This is stable when this in a vertical its the kinetic energy due to If rotation exceeds a certain value. and in so doing would give a considerable amount of kinetic energy.

amount Suppose now that the atom starts with an of kinetic energy well above the critical value. When it gets motion will be very easily disturbed and there will probably be considerable departure from the configuration for steady motion accompanied by a great increase in the rate at which kinetic energy is loss by radiation. i.. The atom and the now emits a much greater number of rays kinetic .e. but be- comes unstable and passes into a different configuration when the kinetic energy sinks below that value. the kinetic energy will decrease in conse- corpuscles quence of the radiation from the rapidly moving but as long as the motion remains . there is a in decrease the great potential energy of the sys- the original configuration tem accompanied by an equal kinetic energy of the corpuscles. is broken up. then.156 ELECTRICITY AND MATTER Let us follow. and it may be thousands of years before the energy approaches the close to this value. energy rapidly approaches the critical value when it reaches this value the crash comes. one which is stable in one configuration of steady motion when the kinetic energy of the corpuscles exceeds a certain value. steady the rate of decrease will be exceedingly slow. increase in the The increase in ÆTHE RF ORCE . the behavior of an atom of this type. the critical value.

corruption responding to the emission of the a rays and the emanation. an atom of the same type i. to discuss the question of other possible sources One source which at once sug- gests itself is external to the radium. and is repeated again for the various radio-active substances. radium ÆTHE RF ORCE ..RADIO-ACTIVE SUBSTANCES 157 the velocity of the corpuscles may cause the disof the atom into two or more systems. This radiation for must be of a very penetrating character. but in a very much shorter time. If the emanation is as the original atom. radium and other rived from an internal source. such as the induced radio-active substance formed out of the emanation. is repeated for the emanation. constitution of the changes in the desirable atom . We might suppose that the radium obtained its energy by absorbing some form o radiation which is passing through all bodies on the surface of the earth. but which is not absorbed to any extent by any but those which are radio-active. We have regarded the energy emitted by radio-active substances as dei.e. as changes of this kind it is have not hitherto been recognized.e. one whose configuration the process for steady motion depends on its kinetic energy. of this energy..

to a slight extent and gave up to the bodies If the direction of the ultra- through which they passed a small fraction of their momentum.158 ELECTRICITY AND MATTER retains its activity when surrounded by thick lead We are familiar or when placed in a deep cellar. with forms of Rontgen rays. absorbed slight absorption. Le Sage supposed that the universe was thronged with exceedingly small particles moving with very high velocities. so that a body A alone in the universe and exposed to ÆTHE RF ORCE . which can produce appreciable effects after passing through several inches of lead. so that the idea of the existence of very peneit trating radiation does not seem so improbable as would have done a few years ago. however. and of rays given out by radium itself. It is interesting to remember that very penetrating radiation was introduced by Le Sage more than a century ago to explain gravitation. the momentum communi- cated by them to the body would not tend to move it in one direction rather than another. mundane corpuscles passing through a body were uniformly distributed. He called these ultra-mundane corpuscles and as- sumed that they were so penetrating that they could pass through masses as large as the sun or the planets without suffering more than a very They were.

in this direction to keep in equilibrium hence. there B in from a second body will shield off the neighborhood of A. to suppose theory. Sage's corpuscles is would re- if. puscles would be sufficient. very penetrating radiation which might escape ÆTHE RF ORCE . however.e. when move B. B is present.RADIO-ACTIVE SUBSTANCES 159 bombardment by Le main at rest . will be attracted to Maxwell pointed out that this transference of momentum from Le Sage's corpuscles to the body through which they were passing involved the loss of kinetic energy by the corpuscles and that if the loss of momen. the kinetic energy lost by the ultra-mundane corif converted into heat. A momentum in this direction as field. i. direction BA the corpuscles moving in the will not receive as much thus. it did when it it was re- alone in the but in the latter case only ceived enough it momentum . The to sufficient to keep was urged an argument against Le Sage's by Maxwell as It is not necessary. that the energy of the corpuscles is transformed into heat we might imagine it transformed into a fact that all bodies are not white hot . the momentum upper hand in the opposite direction will get the so that A will in the direction. B A some of . A B. the gravitating body white hot.. however. tum were account for gravitation.

astronomers believe they have established that it travels with a very much greater velocity. It ought to be mentioned that on this view any changes in gravitation would be propagated with the velocity of light. the loss ÆTHE RF ORCE .1(50 ELECTRICITY AND MATTER from the gravitating body. A "We have seen in the first chapter that waves of electric and magnetic force possess in their direction of propagation. If the absorption of these waves per unit volume depended only upon. whereas. the attraction between the bodies would be directly proportional to the product of their masses. simple calculation will show that the amount of kinetic energy transformed per second in each gram of the gravitating body must be enormously greater than that given out in the same time by one gram of radium. momentum there- we might fore replace Le Sage's corpuscles rays. trating Rontgen Those. the density. would which those to give up momentum to the bodies through they pass. and similar consideration given by Le Sage would show that two bodies would attract each other inversely as the square of the distance between them. if by very peneabsorbed. and was proportional to. As in the case of Le Sage's corpuscles.

trating type than the original in the case of tion of energy rays. must be enormous so great that the energy emitted by radium would be but an exceedingly small fraction of the energy being transformed within it. the absorprays. v being the velocity of light. The reason which induces me to think that the source of the I think that the is from radium energy is in the it is atom of radium itself and not ex- ternal to is. Again. a little reflection will show that he absorption of the rays would not produce To get such attraction the transformed rays must be of a more penegravitational attraction. It may how can this statement be reconciled with the fact ize a transient property. that the radio-activity of substances in all cases in it. as Le Sage's corpuscles. If this energy were transformed into that of rays of the same type as the incident rays. No ÆTHE RF ORCE . which we have been able to local- substance goes on be asked being radio-active for very long. if from these they are the cause of gravitation. From these considerations magnitude of the energy radiated not a valid argument against the energy being derived from radiation.RADIO-ACTIVE SUBSTANCES of IQ ac- momentum by lost the Rontgen rays would be companied by a loss of energy. for each unit of momentum v units of energy would be lost.

the substance which activity. as Rutherford and Soddy have shown in the case of thorium. Take any of the radio-active substances we have described. This is what we should expect on the view that the source of the radio-activity is a change in the atom it is not what we should expect if the source were ex. it is only an exceedingly small fraction of the mass which is at any one time radio-active. the emanations from thorium or radium.162 ELECTRICITY AND MATTER that thorium and radium keep up their activity without any appreciable falling off with time. the ThX. ternal radiation. all produces induced radio- these are active for at the most a few days and then lose this property. and has to be replaced by a fresh supply from the non-radio-active thorium. The answer to this is that. ÆTHE RF ORCE . and that this radio-active portion loses its activity in a few hours.