The paper describes a theory that can be used to explain a mathematical statement of the functioning of an oil system. To arrive at this theory, the rates of contamination of the oil circuits of turbojet engines were investigated extensively, by way of spectrometric oil analyses. The details of the phenomena involved are presented, along with the parameters governing the oil system; a chart for interpreting spectrometric oil analysis data is developed, and the mathematical theory is worked out in detail.There is a relation between the rate of wear of the components of an engine and the rate of contamination of its oil system. By indicating metal particle concentration in the oil, spectrometric oil analysis detects rates of wear high enough to constitute risk of accident due to engine failure. The detection method described takes into account the relation between the rate of contamination and the measured concentration of particles, making it possible to set up contamination rate statistics and to establish component soundness standards.