Before entering into a discussion of the question, which naturally arises at this point, as to how levity and gravity by their two possible ways of interaction - 'sulphurous' or 'saline' - determine the properties of so-called positive and negative electricity, we shall first study the third mode of generating electricity, namely, by electromagnetic induction. Along this way we shall arrive at a picture of the magnetic force which corresponds to the one already obtained of electricity. This will then lead us to a joint study of the nature of electric polarity and magnetic polarity. The discovery of the phenomena we call electromagnetic depended on the possibility of producing continuous electrical processes. This arose with Volta's invention. When it became necessary to find a concept for the process which takes place in an electric conductor between the poles of a galvanic cell, the concept of the 'current', borrowed from hydrodynamics, suggested itself. Ever since then it has been the rule to speak of the existence of a current within an electric circuit; its strength or intensity is measured in terms of a unit named in honour of Ampere. This concept of the current has had a fate typical of the whole relation of human thought to the facts connected with electricity. Long after it had been coined to cover phenomena which in themselves betray no movement of any kind between the electrical poles, other phenomena which do in fact show such movements became known through Crookes's observations. Just as in the case of atomism, they seemed to prove the validity of the preconceived idea of the current. Soon, however, radiant electricity showed properties which contradicted the picture of something flowing from one pole to the other. The cathode rays, for instance, were found to shoot forth into space perpendicularly from the surface of the cathode, without regard to the position of the anode. At the same time Maxwell's hydrodynamic analogy (as our historical survey has shown) led to a view of the nature of electricity by which this very analogy was put out of court. By predicting certain properties of electricity which come to the fore when its poles alternate rapidly, he seemed to bring electricity into close kinship with light. Mathematical treatment then made it necessary to regard the essential energy process as occurring, not from one pole to the other, but at right angles to a line joining the poles (Poynting's vector). This picture, however, satisfactory though it was in the realm of high frequency, failed as a means of describing so-called direct-current processes.