Hume contributed to the theory of ideas by distinguishing impressions from ideas. This refinement of Locke's theory is usually held to be a clarification of Locke's broad use of the term ‘idea,’ and yet the distinction has remained problematic. The bifurcation of the Lockean realm of ideas necessitated an explanation of the relations between the newly named entities. The most basic aspects of this relationship, as presented in the Treatise, are that impressions cause ideas and that ideas represent impressions. In this paper I will argue that Hume's theory cannot accommodate the claim that impressions cause ideas, and that this fundamental inability to trace the causal history of an idea is the source of several other problems in Hume's philosophy of mind.