Do stimulus classes exist before they are tested?

This paper addresses verbal practices that are common when behavior analysis talk about stimulus classes. Specifically, we examine some of the conditions that may set the occasion for saying "Stimulus classes exist prior to the tests that document their formation." We suggest that "stimulus class" should tact behavior that is a function of training and test procedures, not entities that "form" or "exist" in any independent sense. To frame our arguments, we review relevant research findings and suggest descriptive language that is more consistent with behavior analytic traditions.