If the widening scope of mechanization and the diffusion of a machine technology provided a unifying theme for our approach to technological change in the nineteenth century, no such similar key or common denominator is available for the twentieth century. Indeed, the sheer diversity in the sources of technological change is one of the most distinctive features of the twentieth century. It does appear possible, however, to discern a trend underlying this diversity, a trend which can be most conveniently thought of as tracing its origins to the 1860s. That is, an increasing proportion of technological changes have been dependent upon prior advances in systematized knowledge, a knowledge which has brought with it a much more deeply-rooted understanding of the forces of nature and the physical universe. This is a trend which, moreover, begins in the second half of the nineteenth century and therefore the contrast between the nineteenth and twentieth centuries should not be overstated. There was a considerable reliance upon science in the inventive process in some areas in the nineteenth century, and there has been much inventive activity in the twentieth century which remains totally innocent 118of any reliance upon science. 1 The trend, however, is sufficiently pervasive and far advanced that it has altered the whole man vs. environment confrontation in a way which now gives the human agent much more initiative in the adaptive process than he formerly possessed. This shift in the composition of inventive activity away from the empirically-based and toward the newer science-based industries is apparent in the patent statistics. For the period 1916–45 it has been shown that the number of patents granted annually grew much more rapidly in classes dependent upon the application of knowledge from scientific disciplines such as chemistry and physics, whereas the growth in classes dependent upon empirical or practical knowledge and mechanical ingenuity has been much slower. 2 As a more general index of the growing reliance upon inputs of systematized knowledge, it may be pointed out that the number of scientists and technologists has been growing more rapidly than the labor force as a whole since 1870. 3