The central aim of this article is to discuss Russell’s analysis of the notion of cause. In his presidential address to the Aristotelian Society in 1912, Russell put forward several theses on causality in general, and specially on its role in science. He claimed that although vague references to causal laws are often found in the beginnings of science, “in the advanced sciences … the word ‘cause’ never occurs”. Furthermore, Russell maintained that even in philosophy the word ‘cause’ is “so inextricably bound up with misleading associations” that it would be desirable to promote its “complete extrusion from the philosophical vocabulary”. These positions are rendered particularly strong by Russell’s explicit adhesion to the regular sequence view of causation, which he attributed to Hume. Essentially the same opinions were repeated in a series of lectures delivered in Boston two years later. After a systematic exposition of the main theses advanced by Russell on these two occasions, we trace their origin to his general conception of science. Then we examine the substantial changes that Russell’s views on causality underwent several decades later. We remark that these changes seem to be intimately associated with the adoption of a new epistemological position concerning the nature of our knowledge of the external world, involving some clear elements of realism and naturalism.