characterize properties of the medium. The presence of material bodies modifies these properties in ways that depend on the type of substance involved. In the competing view, electricity was explained through sources, the electrons that Helmholtz had postulated, while transmission through space was explained in terms of polarization of a dielectric medium.
In the Treatise Maxwell attempted ". . .a comparison from a philosophical point of view of the results of two methods so completely opposed in their first principles. . ."(p. 502) He did not, however, compare them as competing theories since, he insisted, ". . .we must carefully avoid making any assumption not warranted by experimental evidence. (p. 574. His method, accordingly, was to work from observed phenomena and experimentally established laws, systematize all that was known about electricity and magnetism and, to the degree possible, explain the role of the medium in terms of inferences, rather than assumptions. The result was a long difficult book with a structure so confusing that even such careful students of the Treatise as Hertz and Whittaker admitted a failure to grasp Maxwell's intentions. With a dependence on theoretical assumptions reduced to a minimum there were only three ontologically significant differences that might be tested experimentally: displacement, the localization of energy in the field, and the interpretation of the vector potential (or electro-tonic state) as a momentum.. His development led to suggestions for testing the first two differences.
The Treatise is divided into four parts: electrostatics, electrokinematics, magnetism, and electromagnetism. Each begins in the same general fashion, summarizing the phenomena and established laws and then considering how these are explained by the two approaches. In electrostatics the only significant difference was whether energy resided in the medium or on the surface of conductors. This was untestable, since the methods of calculating energy were mathematically equivalent. In magnetism the competing approaches could be shown to be equivalent.
Maxwell developed in detail one type of situation where the field interpretation could make a difference (Pp. 546-604). In a Lagrangian formulation the energy proper to a current could be broken into three components: T = Tm + Te + Tme, where Tm is the component of kinetic energy due to the motion of conductors, Te represents the flow of electromagnetic energy (or of displacement in the medium), and Tme represents the interaction of kinetic and electromagnetic energy. By a detailed analysis of the corresponding forces Maxwell concluded that three types of effects due to these forces should be experimentally detectable. The first would be a mechanical force causing a properly suspended circular conductor to rotate when there is a sudden change in current. The second involves the effect of an electric moment of inertia acting on a rapidly moving body. The third involves 'mixed forces' acing on a current of electricity. Maxwell built experimental apparatuses to detect the first two effects and found no measurable results. The third should be manifested by the deflection of a galvanometer, but was not detected. Maxwell concluded that either these terms do not exist or are negligibly small.16
In electromagnetism the testable difference between the two approaches concerned the medium and the interpretation of and as characterizing the medium. Maxwell's theory (a causal account) that the forces responsible for the propagation of light reside in the medium led to the conclusion that the dielectric capacity of a transparent medium
16 These tests are analyzed in Everitt, 1975 and in Buchwald, 1985.