Students in Introductory Psychology (N = 75) were administered the Narcissism—Projective (N–P) which asks subjects to describe two Thematic Apperception Test (TAT) cards and two early childhood memories. Protocols were scored using criteria adapted from the DSM-III description of the narcissistic personality disorder. On the basis of these ratings, the high and low narcissists (a = 16) were chosen for further study. These subjects were interviewed by an experienced clinician who was unaware of their performance on the N–P. Results indicate satisfactory interjudge reliability, p < .05, and clinical validity (the association between N–P and interview ratings of narcissism), p < .01. It is concluded that, given the N–P's reliability, validity, and psychodynamic significance, this projective measure is useful, generally, as an instrument in the scientific study of psychoanalysis and, specifically, in the rigorous comparative testing of divergent etiological theories of narcissism.