The past several decades have witnessed a proliferation of research on the dark triad (DT), a set of traits comprising Machiavellianism, narcissism, and psychopathy. The bulk of DT research has been marked by several core assumptions, most notably that each DT construct is a monolithic entity that is clearly separable from its counterpart DT constructs. To examine the tenability of these assumptions, we pooled data from 2 samples of North American community members (ns = 312 and 351) to explore (a) the external validity and profile similarities of DT indicators and (b) the factor structure of the DT. Using general personality dimensions as external criteria, we demonstrated that each DT measure is multidimensional and that subdimensions within DT measures often display sharply different and at times even opposing relations with personality domains; these opposing relations were largely obscured at the total score level adopted in most of the DT literature. In both samples, confirmatory factor analyses and exploratory structural equation models provided no clear support for the traditional tripartite DT structure delineated in the literature. Instead, various aspects of the DT constructs fractionated across a number of factors that represented more basic personality elements (e.g., emotional stability, grandiosity). Taken together, our findings raise serious questions regarding the standard model of DT research and suggest that the questions posed regarding the correlates of DT constructs hinge crucially on the specific DT measure and subdimension examined.