From the archives: The discovery of electrons breaks open the subatomic era

In the August 1901 issue of Popular Science, physicist J. J. Thomson excitedly detailed his methods for discovering the electron.
An image of the 1901 issue of Popular Science Monthly
“On bodies smaller than atoms” (J. J. Thomson, 1901) Popular Science

To mark our 150th year, we’re revisiting the Popular Science stories (both hits and misses) that helped define scientific progress, understanding, and innovation—with an added hint of modern context. Explore the entire From the Archives series and check out all our anniversary coverage here.

As far as human knowledge is concerned, the electron turned 125 on April 30, 2022. Of course, the subatomic particles have been around since shortly after the Big Bang, but here on Earth nobody knew about them until British physicist, J. J. Thomson announced his discovery on April 30, 1897 at the Royal Institution in London. 

In August 1901, Thomson wrote  “On Bodies Smaller Than Atoms” for Popular Science, detailing his discovery and methods. By today’s standards, the piece reads like a hybrid journal article and memoir, capturing his pride and the thrill of discovery. Thomson was awarded the Nobel Prize in Physics for isolating electrons as something fundamental to all atoms.

At the time of Thomson’s finding, no one had ever detected anything smaller than a hydrogen atom (one proton and one electron, no neutron). However, electricity’s ability to flow through materials—coupled, as Thomson cites, with MarieCurie’s radiation experiments and associated electric fields—suggested the possibility. 

Thomson did more than discover electrons; his method, which involved accelerating particles between electrodes, kicked off a new way to study the subatomic world, using accelerators and colliders to smash apart the smallest of the small. By 1911, Ernest Rutherford presented his atomic model, which confirmed Thomson’s electron discovery, but disproved his broader hypothesis that atoms were uniformly distributed (protons paired with electrons). Today, a potpourri of particles smaller than electrons, such as quarks and neutrinos, make up the Standard Model of the universe, developed in the 1970s. The most elusive, perhaps, is the Higgs boson—believed to be the origin of mass for all subatomic particles—first spied in 2012 by physicists at CERN’s Large Hadron Collider. But even the Standard Model has its gaps, like dark matter and antimatter, which, a century on, continues to fuel the quest for bodies smaller than atoms.

“On bodies smaller than atoms” (J. J. Thomson, August 1901)

The masses of the atoms of the various gasses were first investigated about thirty years ago by methods due to Loschmidt, Johnstone Stoney and Lord Kelvin. These physicists, using the principles of the kinetic theory of gasses and making certain assumptions, which it must be admitted are not entirely satisfactory, as to the shape of the atom, determined the mass of an atom of a gas: and when once the mass of an atom of one substance is known the masses of the atoms of all other substances are easily deduced by well-known chemical considerations. 

“The results of these investigations might be thought not to leave much room for the existence of anything smaller than ordinary atoms, for they showed that in a cubic centimeter of gas at atmospheric pressure and at 0° C. there are about 20 million, million, million (2 X 1019) molecules of gas.

Though some of the arguments used to get this result are open to question, the result itself has been confirmed by considerations of quite a different kind. Thus Lord Rayleigh has shown that this number of molecules per cubic centimeter gives about the right value for the optical opacity of the air, while a method, which I will now describe, by which we can directly measure the number of molecules in a gas leads to a result almost identical with that of Loschmidt. This method is founded on Faraday’s laws of electrolysis; we deduce from these laws that the current through an electrolyte is carried by the atoms of the electrolyte, and that all these atoms carry the same charge, so that the weight of the atoms required to carry a given quantity of electricity is proportional to the quantity carried. We know too, by the results of experiments on electrolysis, that to carry the unit charge of electricity requires a collection of atoms of hydrogen which together weigh about 1/10 of a milligram; hence if we can measure the charge of electricity on an atom of hydrogen we see that 1/10 of this charge will be the weight in milligrams of the atom of hydrogen. This result is for the case when electricity passes through a liquid electrolyte. I will now explain how we can measure the mass of the carriers of electricity required to convey a given charge of electricity through a rarefied gas. In this case the direct methods which are applicable to liquid electrolytes cannot be used, but there are other, if more indirect, methods, by which we can solve the problem. The first case of conduction of electricity through gasses we shall consider is that of the so-called cathode rays, those streamers from the negative electrode in a vacuum tube which produce the well-known green phosphorescence on the glass of the tube. These rays are now known to consist of negatively electrified particles moving with great rapidity. Let us see how we can determine the electric charge carried by a given mass of these particles. We can do this by measuring the effect of electric and magnetic forces on the particles. If these are charged with electricity they ought to be deflected when they are acted on by an electric force. It was some time, however, before such a deflection was observed, and many attempts to obtain this deflection were unsuccessful. The want of success was due to the fact that the rapidly moving electrified particles which constitute the cathode rays make the gas through which they pass a conductor of electricity; the particles are thus as it were moving inside conducting tubes which screen them off from an external electric field; by reducing the pressure of the gas inside the tube to such an extent that there was very little gas left to conduct, I was able to get rid of this screening effect and obtain the deflection of the rays by an electrostatic field. The cathode rays are also deflected by a magnet, the force exerted on them by the magnetic field is at right angles to the magnetic force, at right angles also to the velocity of the particle and equal to Hev sin 𝜽 where H is the magnetic force, e the charge on the particle and 𝜽 the angle between H and v. Sir George Stokes showed long ago that, if the magnetic force was at right angles to the velocity of the particle, the latter would describe a circle whose radius is mv/eH (if m is the mass of the particle); we can measure the radius of this circle and thus find m/ve. To find v let an electric force F and a magnetic force H act simultaneously on the particle, the electric and magnetic forces being both at right angles to the path of the particle and also at right angles to each other. Let us adjust these forces so that the effect of the electric force which is equal to Fe just balances that of the magnetic force which is equal to Hev; when this is the case Fe = Hev or v = F. We can thus find v, and knowing from the previous experiment the value of vm/e, we deduce the value of m/e. The value of m/e found in this way was about 10-7, and other methods used by Wiechert, Kaufmann and Lenard have given results not greatly different. Since m/e = 10-7, we see that to carry unit charge of electricity by the particles forming the cathode rays only requires a mass of these particles amounting to one ten thousandth of a milligram while to carry the same charge by hydrogen atoms would require a mass of one-tenth of a milligram.* 

Thus to carry a given charge of electricity by hydrogen atoms requires a mass a thousand times greater than to carry it by the negatively electrified particles which constitute the cathode rays, and it is very significant that, while the mass of atoms required to carry a given charge through a liquid electrolyte depends upon the kind of atom, being, for example, eight times greater for oxygen than for hydrogen atoms, the mass of cathode ray particles required to carry a given charge is quite independent of the gas through which the rays travel and of the nature of the electrode from which they start.

The exceedingly small mass of these particles for a given charge compared with that of the hydrogen atoms might be due either to the mass of each of these particles being very small compared with that of a hydrogen atom or else to the charge carried by each particle being large compared with that carried by the atom of hydrogen. It is therefore essential that we should determine the electric charge carried by one of these particles. The problem is as follows: suppose in an enclosed space we have a number of electrified particles each carrying the same charge, it is required to find the charge on each particle. It is easy by electrical methods to determine the total quantity of electricity on the collection of particles and knowing this we can find the charge on each particle if we can count the number of particles. To count these particles the first step is to make them visible. We can do this by availing ourselves of a discovery made by C. T. R. Wilson working in the Cavendish Laboratory. Wilson has shown that when positively and negatively electrified particles are present in moist dust-free air a cloud is produced when the air is closed by a sudden expansion, though this amount of expansion would be quite insufficient to produce condensation when no electrified particles are present: the water condenses round the electrified particles, and, if these are not too numerous, each particle becomes the nucleus of a little drop of water. Now Sir George Stokes has shown how we can calculate the rate at which a drop of water falls through air if we know the size of the drop, and conversely we can determine the size of the drop by measuring the rate at which it falls through the air, hence by measuring the speed with which the cloud falls we can determine the volume of each little drop; the whole volume of water deposited by cooling the air can easily be calculated, and dividing the whole volume of water by the volume of one of the drops we get the number of drops, and hence the number of the electrified particles. We saw, however, that if we knew the number of particles we could get the electric charge on each particle; proceeding in this way 1 found that the charge carried by each particle was about 6.5 × 10-10 electrostatic units of electricity or 2.17 × 10-20 electro-magnetic units. According to the kinetic theory of gasses, there are 2 × 1019 molecules in a cubic centimeter of gas at atmospheric pressure and at the temperature 0° C.; as a cubic centimeter of hydrogen weighs about 1/11 of a milligram each molecule of hydrogen weighs about 1/ (22 × 1019) milligrams and each atom therefore about 1/(44 × 1019) milligrams and as we have seen that in the electrolysis of solutions one-tenth of 2 milligram carries unit charge, the atom of hydrogen will carry a charge equal to 10/(44 × 1019)= 2.27 × 10-29 electro-magnetic units. The charge on the particles in a gas we have seen is equal to 2.17 × 10-20 units, these numbers are so nearly equal that, considering the difficulties of the experiments, we may feel sure that the charge on one of these gaseous particles is the same as that on an atom of hydrogen in electrolysis. This result has been verified in a different way by Professor Townsend, who used a method by which he found, not the absolute value of the electric charge on a particle, but the ratio of this charge to the charge on an atom of hydrogen and he found that the two charges were equal.

As the charges on the particle and the hydrogen atom are the same, the fact that the mass of these particles required to carry a given charge of electricity is only one-thousandth part of the mass of the hydrogen atoms shows that the mass of each of these particles is only about 1/1000 of that of a hydrogen atom. These particles occurred in the cathode rays inside a discharge tube, so that we have obtained from the matter inside such a tube particles having a much smaller mass than that of the atom of hydrogen, the smallest mass hitherto recognized. These negatively electrified particles, which I have called corpuscles, have the same electric charge and the same mass whatever be the nature of the gas inside the tube or whatever the nature of the electrodes; the charge and mass are invariable. They therefore form an invariable constituent of the atoms or molecules of all gasses and presumably of all liquids and solids.

Nor are the corpuscles confined to the somewhat inaccessible regions in which cathodic rays are found. I have found that they are given off by incandescent metals, by metals when illuminated by ultraviolet light, while the researches of Becquerel and Professor and Madame Curie have shown that they are given off by that wonderful substance the radio-active radium. In fact in every case in which the transport of negative electricity through gas at a low pressure (i.e., when the corpuscles have nothing to stick to) has been examined, it has been found that the carriers of the negative electricity are these corpuscles of invariable mass.

A very different state of things holds for the positive electricity. The masses of the carriers of positive electricity have been determined for the positive electrification in vacuum tubes by Wien and by Ewers, while I have measured the same thing for the positive electrification produced in a gas by an incandescent wire. The results of these experiments show a remarkable difference between the property of positive and negative electrification, for the positive electricity, instead of being associated with a constant mass 1/1000 of that of the hydrogen atom, is found to be always connected with a mass which is of the same order as that of an ordinary molecule. and which, moreover, varies with the nature of the gas in which the electrification is found.

These two results, the invariability and smallness of the mass of the carriers of negative electricity, and the variability and comparatively large mass of the carriers of positive electricity, seem to me to point unmistakably to a very definite conception as to the nature of electricity. Do they not obviously suggest that negative electricity consists of these corpuscles or, to put it the other way, that these corpuscles are negative electricity: and that positive electrification consists in the absence of these corpuscles from ordinary atoms? Thus this point of view approximates very closely to the old one-fluid theory of Franklin; on that theory electricity was regarded as a fluid, and changes in the state of electrification were regarded as due to the transport of this fluid from one place to another. If we regard Franklin’s electric fluid as a collection of negatively electrified corpuscles, the old one-fluid theory will, in many respects, express the results of the new. We have seen that we know a good deal about the ‘electric fluid’; we know that it is molecular or rather corpuscular in character; we know the mass of each of these corpuscles and the charge of electricity carried by it; we have seen too that the velocity with which the corpuscles move can be determined without difficulty. In fact the electric fluid is much more amenable to experiment than an ordinary gas, and the details of its structure are more easily determined.

Negative electricity (i.e., the electric fluid) has mass; a body negatively electrified has a greater mass than the same body in the neutral state; positive electrification, on the other hand, since it involves the absence of corpuscles, is accompanied by a diminution in mass.

An interesting question arises as to the nature of the mass of these corpuscles which we may illustrate in the following way. When a charged corpuscle is moving, it produces in the region around it a magnetic field whose strength is proportional to the velocity of the corpuscle; now in a magnetic field there is an amount of energy proportional to the square of the strength, and thus, in this case, proportional to the square of the velocity of the corpuscle.

Thus if e is the electric charge on the corpuscle and v its velocity, there will be in the region round the corpuscle an amount of energy equal to ½βe2v2? where β is a constant which depends upon the shape and size of the corpuscle. Again if m is the mass of the corpuscle its kinetic energy is ½mv2, and thus the total energy due to the moving electrified corpuscle is ½(m + βe2)v2,so that for the same velocity it has the same kinetic energy as a non-electrified body whose mass is greater than that of the electrified body by βe2. Thus a charged body possesses in virtue of its charge, as I showed twenty years ago, an apparent mass apart from that arising from the ordinary matter in the body. Thus in the case of these corpuscles, part of their mass is undoubtedly due to their electrification, and the question arises whether or not the whole of their mass can be accounted for in this way. I have recently made some experiments which were intended to test this point; the principle underlying these experiments was as follows: if the mass of the corpuscle is the ordinary “mechanical mass, then, if a rapidly moving corpuscle is brought to rest by colliding with a solid obstacle, its kinetic energy being resident in the corpuscle will be spent in heating up the molecules of the obstacle in the neighborhood of the place of collision, and we should expect the mechanical equivalent of the heat produced in the obstacle to be equal to the kinetic energy of the corpuscle. If, on the other hand, the mass of the corpuscle is “electrical,” then the kinetic energy is not in the corpuscle itself, but in the medium around it, and, when the corpuscle is stopped, the energy travels outwards into space as a pulse confined to a thin shell traveling with the velocity of light. I suggested some time ago that this pulse forms the Rontgen rays which are produced when the corpuscles strike against an obstacle. On this view, the first effect of the collision is to produce Rontgen rays and thus, unless the obstacle against which the corpuscle strikes absorbs all these rays, the energy of the heat developed in the obstacle will be less than the energy of the corpuscle. Thus, on the view that the mass of the corpuscle is wholly or mainly electrical in its origin, we should expect the heating effect to be smaller when the corpuscles strike against a target permeable by the Rontgen rays given out by the tube in which the corpuscles are produced than when they strike against a target opaque to these rays. I have tested the heating effects produced in permeable and opaque targets, but have never been able to get evidence of any considerable difference between the two cases. The differences actually observed were small compared with the total effect and were sometimes in one direction and sometimes in the opposite. The experiments, therefore, tell against the view that the whole of the mass of a corpuscle is due to its electrical charge. The idea that mass in general is electrical in its origin is a fascinating one, although it has not at present been reconciled with the results of experience.

The smallness of these particles marks them out as likely to afford a very valuable means for investigating the details of molecular structure, a structure so fine that even waves of light are on far too large a scale to be suitable for its investigation, as a single wavelength extends over a large number of molecules. This anticipation has been fully realized by Lenard’s experiments on the obstruction offered to the passage of these corpuscles through different substances. Lenard found that this obstruction depended only upon the density of the substance and not upon its chemical composition or physical state. He found that, if he took plates of different substances of equal areas and of such thicknesses that the masses of all the plates were the same, then, no matter what the plates were made of, whether of insulators or conductors, whether of gasses, liquids or solids, the resistance they offered to the passage of the corpuscles through them was the same. Now this is exactly what would happen if the atom of the chemical elements were aggregations of a large number of equal particles of equal mass; the mass of an atom being proportional to the number of these particles contained in it and the atom being a collection of such particles through the interstices between which the corpuscle might find its way. Thus a collision between a corpuscle and an atom would not be so much a collision between the corpuscle and the atom as a whole, as between a corpuscle and the individual particles of which the atom consists; and the number of collisions the corpuscle would make, and therefore the resistance it would experience, would be the same if the number of particles in unit volume were the same, whatever the nature of the atoms might be into which these particles are aggregated. The number of particles in unit volume is however fixed by the density of the substance and thus on this view the density and the density alone should fix the resistance offered by the substance to the motion of a corpuscle through it; this, however, is precisely Lenard’s result, which is thus a strong confirmation of the view that the atoms of the elementary substances are made up of simpler parts all of which are alike. This and similar views of the constitution of matter have often been advocated; thus in one form of it, known as Prout’s hypothesis, all the elements were supposed to be compounds of hydrogen. We know, however, that the mass of the primordial atom must be much less than that of hydrogen. Sir Norman Lockyer has advocated the composite view of the nature of the elements on spectroscopic grounds, but the view has never been more boldly stated than it was long ago by Newton who says:

“The smallest particles of matter may cohere by the strongest attraction and compose bigger particles of weaker virtue and many of these may cohere and compose bigger particles whose virtue is still weaker and so on for divers succession, until the progression ends in the biggest particles on which the operations in Chemistry and the colours of natural bodies depend and which by adhering compose bodies of a sensible magnitude.”

The reasoning we used to prove that the resistance to the motion of the corpuscle depends only upon the density is only valid when the sphere of action of one of the particles on a corpuscle does not extend as far as the nearest particle. We shall show later on that the sphere of action of a particle on a corpuscle depends upon the velocity of the corpuscle, the smaller the velocity the greater being the sphere of action and that if the velocity of the corpuscle falls as low as 107 centimeters per second, then, from what we know of the charge on the corpuscle and the size of molecules, the sphere of action of the particle might be expected to extend further than the distance between two particles and thus for corpuscles moving with this and smaller velocities we should not expect the density law to hold. 

Existence of free corpuscles or negative electricity in metals

In the cases hitherto described the negatively electrified corpuscles had been obtained by processes which require the bodies from which the corpuscles are liberated to be subjected to somewhat exceptional treatment. Thus in the case of the cathode rays the corpuscles were obtained by means of intense electric fields. in the case of the incandescent wire by great heat, in the case of the cold metal surface by exposing this surface to light. The question arises whether there is not to some extent, even in matter in the ordinary state and free from the action of such agencies, a spontaneous liberation of those corpuscles a kind of dissociation of the neutral molecules of the substance into positively and negatively electrified parts, of which the latter are the negatively electrified corpuscles.

Let us consider the consequences of some such effect occurring in a metal, the atoms of the metal splitting np into negatively electrified corpuscles and positively electrified atoms and these again after a time re-combining to form neutral system. When things have got into a steady state the number of corpuscles re-combining in a given time will be equal to the number liberated in the same time. There will thus be diffused through the metal swarms of these corpuscles, these will be moving about in all directions like the molecules of a gas and, as they can gain or lose energy by colliding with the molecule of the metal, we should expect by the kinetic theory of gasses that they will acquire such an average velocity that the mean kinetic energy of a corpuscle moving about in the metal is equal to that possessed by a molecule of a gas at the temperature of the metal; this would make the average velocity of the corpuscles at 0° C. about 107 centimeters per second. This swarm of negatively electrified corpuscles when exposed to an electric force will be sent drifting along in the direction opposite to the force; this drifting of the corpuscles will be an electric current, so that we could in this way explain the electrical conductivity of metals.

The amount of electricity carried across unit area under a given electric force will depend upon and increase with (1) the number of free corpuscles per unit volume of the metal, (2) the freedom with which these can move under the force between the atoms of the metal; the latter will depend upon the average velocity of these corpuscles, for if they are moving with very great rapidity the electric force will have very little time to act before the corpuscle collides with an atom, and the effect produced by the electric force annulled. Thus the average velocity of drift imparted to the corpuscles by the electric field will diminish as the average velocity of translation, which is fixed by the temperature, increases. As the average velocity of translation increases with the temperature, the corpuscles will move more freely under the action of an electric force at low temperatures than at high, and thus from this cause the electrical conductivity of metals would increase as the temperature diminishes. In a paper presented to the International Congress of Physics at Paris in the autumn of last year, I described a method by which the number of corpuscles per unit volume and the velocity with which they moved under an electric force can be determined. Applying this method to the case of bismuth, it appears that at the temperature of 20° C, there are about as many corpuscles in a cubic centimeter as there are molecules in the same volume of a gas at the same temperature and at a pressure of about ¼ of an atmosphere, and that the corpuscles under an electric field of 1 volt per centimeter would travel at the rate of about 70 meters per second. Bismuth is at present the only metal for which the data necessary for the application of this method exists, but experiments are in progress at the Cavendish Laboratory which it is hoped will furnish the means for applying the method to other metals. We know enough, however, to be sure that the corpuscles in good conductors, such as gold, silver or copper, must be much more numerous than in bismuth, and that the corpuscular pressure in these metals must amount to many atmospheres. These corpuscles increase the specific heat of a metal and the specific heat gives a superior limit to the number of them in the metal.

An interesting application of this theory is to the conduction of electricity through thin films of metal. Longden has recently shown that when the thickness of the film falls below a certain value, the specific resistance of the film increases rapidly as the thickness of the film diminishes. This result is readily explained by this theory of metallic conduction, for when the film gets so thin that its thickness is comparable with the mean force path of corpuscle the number of collisions made by a corpuscle in a film will be greater than in the metal in bulk, thus the mobility of the particles in the film will be less and the electrical resistance consequently greater.

The corpuscles disseminated through the metal will do more than carry the electric current, they will also carry heat from one part to another of an unequally heated piece of metal. For if the corpuscles in one part of the metal have more kinetic energy than those in another, then, in consequence of the collisions of the corpuscles with each other and with the atoms, the kinetic energy will tend to pass from those places where it is greater to those where it is less, and in this way heat will flow from the hot to the cold parts of the metal, as the rate with which the heat is carried will increase with the number of corpuscles and with their mobility, it will be influenced by the same circumstances as the conduction of electricity, so that good conductors of electricity should also be good conductors of heat. If we calculate the ratio of the thermal to the electric conductivity on the assumption that the whole of the heat is carried by the corpuscles we obtain a value which is of the same order as that found by experiment.

Weber many years ago suggested that the electrical conductivity of metals was due to the motion through them of positively and negatively electrified particles, and this view has recently been greatly extended and developed by Riecke and by Drude, the objection to any electrolytic view of the conduction through metals is that, as in electrolysis, the transport of electricity involves the transport of matter, and no evidence of this has been detected, this objection does not apply to the theory sketched above, as on this view it is the corpuscles which carry the current, these are not atoms of the metal, but very much smaller bodies which are the same for all metals.

It may be asked if the corpuscles are disseminated through the metal and moving about in it with an average velocity of about 107 centimeters per second, how is it that some of them do not escape from the metal into the surrounding air? We must remember, however, that these negatively electrified corpuscles are attracted by the positively electrified atoms and in all probability by the neutral atoms as well, so that to escape from these attractions and get free a corpuscle would have to possess a definite amount of energy, if a corpuscle had less energy than this then, even though projected away from the metal, it would fall back into it after traveling a short distance. When the metal is at a high temperature, as in the case of the incandescent wire, or when it is illuminated by ultraviolet light some of the corpuscles acquire sufficient energy to escape from the metal and produce electrification in the surrounding gas. We might expect too that, if we could charge a metal so highly with negative electricity, that the work done by the electric field on the corpuscle in a distance not greater than the sphere of action of the atoms on the corpuscles was greater than the energy required for a corpuscle to escape, then, the corpuscles would escape and negative electricity stream from the metal. In this case the discharge could be effected without the participation of the gas surrounding the metal and might even take place in an absolute vacuum, if we could produce such a thing. We have as yet no evidence of this kind of discharge, unless indeed some of the interesting results recently obtained by Earhart with very short sparks should be indications of an effect of this kind.

A very interesting case of the spontaneous emission of corpuscles is that of the radio-active substance radium discovered by M. and Madame Curie. Radium gives out negatively electrified corpuscles which are deflected by a magnet. Becquerel has determined the ratio of the mass to the charge of the radium corpuscles and finds it is the same as for the corpuscles in the cathode rays. The velocity of the radium corpuscles is, however, greater than any that has hitherto been observed for either cathode or Lenard rays: being, as Becquerel found, as much as 2 X 1010 centimeters per second or two-thirds the velocity of light. This enormous velocity explains why the corpuscles from radium are so very much more penetrating than the corpuscles from cathode or Lenard rays; the difference in this respect is very striking, for while the latter can only penetrate solids when they are beaten out into the thinnest films, the corpuscles from radium have been found by Curie to be able to penetrate a piece of glass 3 millimeters thick. To see how an increase in the velocity can increase the penetrating power, let us take as an illustration of a collision between the corpuscle and the particles of the metal the case of a charged corpuscle moving past an electrified body; a collision may be said to occur between these when the corpuscle comes so close to the charged body that its direction of motion after passing the body differs appreciably from that with which it started. A simple calculation shows that the deflection of the corpuscle will only be considerable when the kinetic energy, with which the corpuscle starts on its journey towards the charged body is not large compared with the work done by the electric forces on the corpuscle in its journey to the shortest distance from the charged body. If d is the shortest distance, e and e’ the charge of the body and corpuscles, the work done is ee’/d; while if m is the mass and v the velocity with which the corpuscle starts the kinetic energy to begin with is ½mv2; thus a considerable deflection of the corpuscle, i.e., a collision will occur only when ee’/d is comparable with ½mv2; and d the distance at which a collision occurs, will vary inversely as v2. As d is the radius of the sphere of action for collision and as the number of collisions is proportional to the area of a section of this sphere, the number of collisions is proportional to d2, and therefore varies inversely as v4. This illustration explains how rapidly the number of collisions and therefore the resistance offered to the motion of the corpuscles through matter diminishes as the velocity of the corpuscles increases, so that we can understand why the rapidly moving corpuscles from radium are able to penetrate substances which are nearly impermeable to the more slowly moving corpuscles from cathode and Lenard rays.

Cosmical effects produced by corpuscles

As a very hot metal emits these corpuscles it does not seem an improbable hypothesis that they are emitted by that very hot body, the sun. Some of the consequences of this hypothesis have been developed by Paulsen, Birkeland and Arrhenius who have developed a theory of the Aurora Borealis from this point of view. Let us suppose that the sun gives out corpuscles which travel out through interplanetary space; some of these will strike the upper regions of the Earth’s atmosphere and will then or even before then, come under the influence of the Earth’s magnetic field. The corpuscles when in such a field, will describe spirals round the lines of magnetic force; as the radii of these spirals will be small compared with the height of the atmosphere; we may for our present purpose suppose that they travel along the lines of the Earth’s magnetic force. Thus the corpuscles which strike the Earth’s atmosphere near the equatorial regions where the lines of magnetic force are horizontal will travel horizontally, and will thus remain at the top of the atmosphere where the density is so small that but little luminosity is caused by the passage of the corpuscles through the gas; as the corpuscles travel into higher latitudes where the lines of magnetic force dip, they follow these lines and descend into the lower and denser parts of the atmosphere, where they produce luminosity, which on this view is the Aurora.

As Arrhenius has pointed out the intensity of the Aurora ought to be a maximum at some latitude intermediate between the pole and the equator, for, though in the equatorial regions the rain of corpuscles from the sun is greatest, the Earth’s magnetic force keeps these in such highly rarefied gas that they produce but little luminosity, while at the pole, where the magnetic force would pull them straight down into the denser air, there are not nearly so many corpuscles; the maximum luminosity will therefore be somewhere between these places. Arrhenius has worked out this theory of the Aurora very completely and has shown that it affords a very satisfactory explanation of the various periodic variations to which it is subject.

As a gas becomes a conductor of electricity when corpuscles pass through it, the upper regions of the air will conduct, and when air currents occur in these regions, conducting matter will be driven across the lines of force due to the Earth’s magnetic field, electric currents will be induced in the air, and the magnetic force due to these currents will produce variations in the Earth’s magnetic field. Balfour Stewart suggested long ago that the variation on the Earth’s magnetic field was caused by currents in the upper regions of the atmosphere, and Schuster has shown, by the application of Gauss’ method, that the seat of these variations is above the surface of the Earth.

The negative charge in the Earth’s atmosphere will not increase indefinitely in consequence of the stream of negatively electrified corpuscles coming into it from the sun, for as soon as it gets negatively electrified it begins to repel negatively electrified corpuscles from the ionized gas in the upper regions of the air, and a state of equilibrium will be reached when the Earth has such a negative charge that the corpuscles driven by it from the upper regions of the atmosphere are equal in number to those reaching the Earth from the sun. Thus, on this view, interplanetary space is thronged with corpuscular traffic, rapidly moving corpuscles coming out from the sun while more slowly moving ones stream into it.

In the case of a planet which, like the moon, has no atmosphere there will be no gas for the corpuscles to ionize, and the negative electrification will increase until it is so intense that the repulsion exerted by it on the corpuscles is great enough to prevent them from reaching the surface of the planet.

Arrhenius has suggested that the luminosity of nebulae may not be due to high temperature, but may be produced by the passage through their outer regions of the corpuscles wandering about in space, the gas in the nebulae being quite cold. This view seems in some respects to have advantages over that which supposes the nebulae to be at very high temperatures. These and other illustrations, which might be given did space permit, seem to render it probable that these corpuscles may play an important Dart in cosmical as well as in terrestrial physics.

*Professor Schuster in 1889 was the first to apply the method of the magnetic deflection of the discharge to get a determination of the value of m/e; he found rather widely separated limiting values for this quantity and came to the conclusion that it was of the same order as in electrolytic solutions, the result of the method mentioned above as well as those of Wiechert, Kaufmann and Leonard make it very much smaller.

The cover of August 1901’s Popular Science Monthly.

Some text has been edited to match contemporary standards and style.