In this sequel to his "Foundations of Metaphysics in Science", Harris develops a new theory of scientific method. Harris challenges the empiricist approach, criticising its presuppositions as internally incoherent and incompatible with actual scientific practice. Looking to C S Peirce and R G Collingwood as precursors, Harris argues that the actual method of thinking employed by scientists is neither inductive nor deductive. Rather, scientific methodology is constructive of systems, not built up from particular, theory-neutral observations of 'matters of fact', but always developed from earlier hypotheses, which, in the course of application, have proved inconsistent. Harris maintains that the advance of science is therefore dialectical in a manner that recent reformers of the empiricist doctrine (such as Thomas Kuhn and Karl Popper) have missed.