September 4, 1903.] 



SCIENCE. 



313 



other words it has apparently a greater inertia 

 mass, than the same body without charge, and 

 the electrical part of the mass is greater and 

 greater the smaller the size of the body. The 

 idea, which is fast gaining strength, that 

 atoms are aggregates of extremely small 

 charged particles, corpuscles as J. J. Thom- 

 son calls them, has given credit to the hj-poth- 

 esis that inertia is wholly electrical in char- 

 acter, inasmuch as an atom is known not tn 

 be small enough, assuming it to be homo- 

 geneous, to permit of its inertia being at- 

 tributed wholly to its electric charge. The 

 word electron seems to be most acceptable as 

 a name for these ultimate corpuscles of dis- 

 embodied electric charge. 



The ' mass ' of an electron is not fixed in 

 value but it increases with velocity. In fact 

 the very rapidly moving electrons (95 per 

 cent, of the velocity of light) in Becquerel 

 rays have greater mass than the more slowly 

 moving ions in cathode rays, and measure- 

 ments of mass and velocity of the electrons 

 in cathrode rays and in Becquerel rays are 

 in accord with the theoretical law of increase 

 of mass with velocity. When an electron is 

 accelerated in the direction of its motion 

 its mass is different (longitudinal mass) from 

 when it is accelerated at right angles, to its 

 <iirection of motion (transversal mass). The 

 difference between longitudinal and trans- 

 versal mass is sufficiently great numerically, 

 when the velocity is great, to be an essential 

 element in the established correspondence be- 

 tween theory and observation above men- 

 tioned. 



The inertia or mass of the electron is purely 

 electromagnetic in character; and if atoms 

 and ynolecitles are 7nere aggregates of electrons 

 it folloirs that all mass and inertia are electro- 

 magnetic in character. 



Some very curious laws of motion arise 

 from the electron theory. When velocities 

 and accelerations are very small Newton's 

 laws of motion apply as very close appro.xima- 

 tions. When velocities or accelerations are 

 large Newton's laws of motion fail utterly. 

 One point of failure is the above-mentioned 

 distinction between longitudinal and trans- 

 verse mass — mass is not a scalar quantity 



but a so-called linear vector function ; other 

 curious points in the motion of electrons are 

 the following: 



Uniform motion in a straight line once 

 established is steady or permanent. Uniform 

 motion of an electron in a circular orbit, on 

 the other hand, is accompanied by continuous 

 radiation of energy, so that such motion must 

 eventually cease if it is not maintained by 

 some external agent. Thus two electrons 

 rotating about each other after the manner of 

 planet and satellite slowly radiate energy, 

 and their motion is affected very much as if 

 they were in a resisting medium. This mo- 

 tion of electrons as satellites is the funda- 

 mental hypothesis in the theory of the Zeeman 

 effect. 



A continuously accelerated electron stores 

 an increasing amount of kinetic energy which 

 can be regained by stopping the electron, 

 and in addition it radiates energy continu- 

 ously, which energy is lost. If the accelera- 

 tion is very great this radiation may be con- 

 siderable. Therefore, there is a drag upon 

 the electron which is something like viscous 

 friction but which depends upon acceleration 

 and not on velocity. 



An electron under way will, if quickly stop- 

 ped and released, start up again of itself and 

 move with diminished velocity in the original 

 direction. 



When an electron gains a certain velocity 

 under the action of an accelerating force and 

 the force suddenly ceases to act, then the elec- 

 tron loses some of its velocity and settles down 

 to a slightly decreased uniform velocity. 



An electron started quickly and kept mov- 

 ing at uniform speed must be acted upon, of 

 course, by a great force to produce the quick 

 start, and also by a lasting force which slowly 

 drops to zero. 



The kinetic energj' of an electron is not 

 proportional to the square of its velocity. 

 This is another way of stating the fact which 

 lies at the foundation of the variation of 

 mass with velocity, and of the distinction be- 

 tween longitudinal and transverse mass. 

 Thus, the onl.v case in which an added velocity 

 )" in the direction of the (/-axis of reference 

 iiicans the same amount of work done whether 



