514 Prof. J. J. Thomson on the Relation between the 



vertical axis, the behaviour of the system, if we make the 

 framework rotate in the direction in which the fly-wheel is 

 spinning, will be very different from its behaviour when we 

 make the framework rotate in the opposite direction. If the 

 framework rotates in the same direction as the fly-wheel, the 

 axis of the wheel remains vertical, and the whole system 

 goes on rotating quietly until stopped by friction. If, 

 however, we attempt to set the framework rotating in the 

 opposite direction to the fly-wheel, the axis of the wheel begins 

 to wobble about, the disturbance gets more violent until finally 

 the fly-wheel topples right over ; then the fly-wheel is rotating 

 in the same direction as the framework, and the rotation goes 

 smoothly on. If the axis of the fly-wheel were held in its 

 original position by springs, then in the first case these 

 springs would not be stretched ; but in the second case they 

 would, and the axis of the fly-wheel would take up a position 

 inclined to the vertical, the angle it made with the vertical 

 depending on the stiffness of the springs and the moment of 

 momentum of the fly-wheel. Thus in the second case the 

 attempt to make the framework of the gyrostat rotate would 

 be accompanied by the storing up of potential energy in the 

 system due to the stretching of the springs, while there would 

 be no such storage of potential energy in the first case. 



Suppose, now, that an atom of hydrogen possesses a struc- 

 ture analogous to this gyroscopic system with springs, the 

 gyrostats rotating in the same direction as the fluid in a 

 Faraday tube leaving the atom. Then, since a charge of 

 negative electricity on the atom implies the incidence of a 

 tube on the atom, when the hydrogen atom is charged nega- 

 tively the rotation in the tube connected with the atom is in 

 the opposite direction to that of its gyrostats. The negative 

 charge will thus cause an increase in the potential energy of 

 the atom, whereas a positive charge when the rotation in the 

 tube is in the same direction as that of the gyrostats does not 

 cause any such increase. Thus the internal potential energy 

 of the hydrogen atom will ceteris paribus be greater when it 

 has a negative charge than when it has a positive one. In 

 the case of a strongly electronegative element such as chlorine, 

 we suppose that the gyrostats in the atom are rotating in the 

 opposite direction to those in the hydrogen atom, i. e. that in 

 the chlorine atom the rotation in the gyrostats is in the oppo- 

 site direction to that of the liquid in a Faraday tube leaving 

 the atom : thus the chlorine atom will, cceteris paribus, have 

 more internal potential energy when charged with positive 

 electricity than it has when charged with negative. 



The existence of the property conferred by these gyrostats 



