Recent studies of Kant's views on science by Friedman (1986, 1992), Kitcher (1983, 1986), Brittan (1978, 1986), Harper (1989), Morrison (1989), and Butts (1979, 1986a&c, 1990) - all of us heavily dependent upon Buchdahl (1969), have revealed that Kant introduces a new form of the old rationalist theme: to build a science is to build asystem)In one respect Kant simply imitates the Leibnizian background (reinforced by rationalists since the time of Plato): in the absence of an understanding of the whole, one cannot hope to understand the parts. However, unlike Leibniz, Kant's philosophy failed to give him the metaphysical assurance that such a feat of mental gymnastics can actually be carried out. Instead, Kant derived his thoughts about systematization from the study of the best science of his day — the Newtonian synthesis — thoughts that led him to replace alleged metaphysical knowledge with certain postulates of rational methodology.