August 26, 1909] 



NA TURE 



259 



all dhecUons at considerable speeds. For a long time 

 this effect, known as the Brownian movement, was ascribed 

 to inequalities in the temperature of the solution. This 

 was disproved by a number of subsequent investigations, 

 and especially by those of Gouy, who showed that the 

 movement was spontaneous and continuous, and was ex- 

 hibited by very small particles of whatever kind when 

 immersed in a fluid medium. The velocity of agitation 

 increased with decrease of diameter of the particles and 

 increased with temperature, and was dependent on the 

 viscosity of the surrounding fluid. With the advent of the 

 ultra-microscope it has been possible to follow the move- 

 ments with more certainty and to e.xperiment with much 

 smaller particles. E.xner and Zsigmondy have determined 

 the mean velocity of particles of known diameter in 

 various solutions, while Svedberg has devised an ingenious 

 method of determining the mean free path and the average 

 velocity of particles of different diameter. The experiments 

 of Ehrenhaft in 1907 showed that the Brownian movement 

 was not confined to liquids, but was exhibited far more 

 markedly by small particles suspended in gases. By 

 passing an arc discharge between silver poles he produced 

 a fine dust of silver in the air. When examined by means 

 of the ultra-microscope the suspended particles exhibited 

 the characteristic Brownian movement, with the difference 

 that the mean free path for particles of the same size 

 •.vas much greater in gases than in liquids. 



The particles exhibit in general the character of the 

 motion which the kinetic theory ascribes to the molecules 

 themselves, although even the smallest particles examined 

 have a mass which is undoubtedly very large compared 

 with that of the molecule. The character of the Brownian 

 movement irresistibly impresses the observer with the idea 

 that the particles are hurled hither and thither by me 

 action of forces resident in the solution, and that these 

 can only arise from the continuous and ceaseless move- 

 ment of the invisible molecules of which the fluid is com- 

 posed. Smctluchowski and Einstein have suggested ex- 

 planations which are based on the kinetic theory, and 

 there is a fair agreement between calculation and experi- 

 ment. Strong additional confirmation of this view has been 

 supplied by the very recent experiments of Perrin (iqoq). 

 He obtained an emulsion of gamboge in water which 

 consisted of a great number of spherical particles nearly 

 of the same size, which showed the characteristic 

 Brownian movement. The particles settled under gravity, 

 and when equilibrium was set up the distribution of these 

 particles in layers at dilTerent heights was determined 

 by counting the particles with a microscope. The number 

 was found to diminish from the bottom of the vessel 

 upwards according to an exponential law — j.t'. according 

 to the same law as the pressure of the atmosphere 

 diminishes from the surface of the earth. In this case, 

 however, on account of the great mass of the particles, 

 their distribution was confined to a region only a fraction 

 of a millimetre deep. In a particular experiment the 

 number of particles per unit volume decreased to half in 

 a distance of o 03S millimetre, while the corresponding 

 distance in our atmosphere is about 6000 metres. From 

 measurements of the diameter and weight of each particle, 

 Perrin found that, within the limit of experimental error, 

 the law of distribution with height indicated that each 

 small particle had the same average kinetic energy of 

 movement as the molecules of the solutions in which they 

 were suspended ; in fact, the particles in suspension behaved 

 in all resoects like molecules of verv high molecular 

 weight. This is a very important result, for it indicates 

 that the law of equipartition of energv among molecules 

 of different masses, which is an important deduction from 

 the kinetic theory, holds, at any rate very approximately, 

 for a distribution of particles in a medium the masses and 

 dimensions of which are exceedinglv large compared with 

 that of the molecules of the medium. Whatever may 

 prove to be the exact explanation of this phenomenon, 

 there can be little doubt that it results from the movement 

 of the molecules of the solution, and is thus a strikin." 

 if somewhat indirect proof of the general correctness of 

 thp kinetic theory of matter. 



From recent work in radio-activitv we mav take a 

 second illustration which is novel and far more direct. It 

 is w-ell known that the a rnvs of radium are deflected by 

 NO. 2078, VOL. 81] 



both magnetic and electric fields. It may be concluded 

 from this evidence that the radiation is corpuscular in 

 character, consisting of a stream of positively charged 

 particles projected from the radium at a very high 

 velocity. From the measurements of the deflection of tlie 

 rays in passing through magnetic and electric fields the 

 ratio elm of the charge carried by the particle to its mass 

 has been determined, and the magnitude of this quantity 

 indicates that the particle is of atomic dimensions. 



Rutherford and Geiger have recently developed a direct 

 method of shov/ing that this radiation is, as the other 

 evidence indicated, discontinuous, and that it is possible 

 to detect by a special electric method the passage of a 

 single a particle into a suitable detecting vessel. The 

 entrance of an a particle through a small opening was 

 marked by a sudden movement of the needle of the electro- 

 meter which was used as a measuring instrument. In 

 this way, by counting the number of separate impulses 

 communicated to the electrometer needle, it was possible 

 to determine by direct counting the number of a particles 

 expelled per second from one gram of radium. But we 

 can go further and confirm the result by counting the 

 number of a. particles by an entirely distinct method. Sir 

 William Crookes has shown that when the a. rays are 

 allowed to fall upon a screen of phosphorescent zmc 

 sulphide, a number of brilliant scintillations are observed. 

 It appears as if the impact of each a particle produced 

 a visible flash of light where it struck the screen. Using 

 suitable screens, the number of scintillations per second 

 on a given area can be counted by means of a microscope. 

 It has been shown that the number of scintillations deter- 

 mined in this way is equal to the number of impinging 

 o particles when counted by the electric method. This 

 shows that the impact of each o particle on the zinc 

 sulphide produces a visible scintillation. There are thus 

 two distinct methods — one electrical, the other optical — • 

 for detecting the emission of a single o particle from 

 radium. The next question to consider is the nature of 

 the a particle itself. The general evidence indicates that 

 the a particle is a charged atom of helium, and this con- 

 clusion was decisively verified by Rutherford and Royds 

 bv showing that helium appeared in an exhausted space 

 into which the a particles were fired. The helium, which 

 is produced by radium, ;s due to the accumulated a. particles 

 which are so continuously expelled from it. If the rate 

 of production of helium from radium is measured, we thus 

 have a means of determining directly how many a particles 

 are required to form a given volume of helium gas. This 

 rate of production has recently been measured accurately 

 bv Sir James Dewar. He has informed me that his final 

 measurements show that one gram of radium in radio- 

 active equilibrium produces 046 cubic millimetre of helium 

 per day, or 5-32x10-° cubic millimetres per second. Now 

 from the direct counting experiments it is known that 

 ij-bxio" o particles arc shot out per second from one 

 gram of radium in equilibrium. Consequently it requires 

 2-56xio"' a particles to form one cubic centimetre of 

 helium gas at standard pressure and temperature. 



From other lines of evidence it is known that all the 

 a particles, from whatever source, are identical in mass 

 and constitution. It is not, then, unreasonable to suppose 

 that the a particle, which exists as a separate entity in 

 its flight, can exist also as a separate entity when the 

 a particles are collected together to form a measurable 

 volume of helium gas, or, in other words, that the 

 a particle on losing its charge becomes the fundamental 

 imit or atom of helium. In the case of a monatomic gas 

 like helium, where the atom and molecule are believed 

 to be identical, no difficulty of deduction arises from the 

 possible combination of two or more atoms to form .1 

 complex molecule. 



We consequently conclude from these experiments that 

 one cubic centimetre of helium at standard pressure and 

 temperature contains 256 x 10" atoms. Knowing the 

 density of helium, it at once follows that each atom of 

 helium has a mass of fi-Sxio-"* grams, and that the 

 average distance apart of the molecules in the gaseous 

 state at standard pressure and temperature is 34X10-' 

 centimetres. 



The above result can be confirmed in a different wav. 

 It is known that the value of ehn for the o particle is 



