The ‘Wave-Band’ Theory of Wireless Transmission

Abstract

MY friend Sir Ambrose Fleming raises an interesting question in his admirably clear article on page 92 of NATURE of Jan. 18, no less a question than whether a mathematical alternative does or does not invariably correspond with some physical reality. I am inclined to think that it does. For example, there is something rotatory in a magnetic field, and whether the rotation of a plane of polarisation may be properly expressed as an acceleration of one circular component and the retardation of another, we have no better mode of expression until we know more. So also I think that a sinuous wave of fluctuating amplitude may be rightly and exactly represented as if it were a band of neighbouring frequencies. This is not obvious but I would remind Sir Ambrose Fleming that the complete solution of the relevant differential equation for forced vibrations contains not only a simply periodic e^ipt term, but one with an evanescent exponential amplitude like e−^kt as well and these latter periodic effects—depending as they do on the natural frequency of the receiver circuit (or on its range of possible frequencies)—though they rapidly die away—are of influence at the beginning and end of a wave series. Dr. Eccles tells me that those initial and final effects are called ‘transients’ by electrical engineers and that is a good name for them.