Metaphysicians speak of laws of nature in terms of necessity and universality; scientists do so in terms of symmetry and invariance. This book argues that no metaphysical account of laws can succeed. The author analyses and rejects the arguments that there are laws of nature, or that we must believe that there are. He argues that we should discard the idea of law as an inadequate clue to science. After exploring what this means for general epistemology, the book develops the empiricist view of science as a construction of models to represent the phenomena. Concepts of symmetry, transformation, and invariance illuminate the structure of such models. A central role is played in science by symmetry arguments, and it is shown how these function also in the philosophical analysis of probability. The advocated approach presupposes no realism about laws or necessities in nature.