There exists a long history of investigation regarding the relationship between personality and affective disorders, specifically, bipolar disorder. A better understanding of this relationship can lead to more accurate diagnoses, improved prognosis and possible improvement of quality of life for those with the disorder. One challenge in this area of study is disentangling symptoms from personality traits, even with the use of patient groups in remission. One approach to this problem is to include in the study a group of relatives of individuals with mood disorders to capture those considered “at-risk,” but without the occurrence of the illness in that group. The NEO-PI-R is an empirically-based, well-established personality assessment questionnaire that measures the five-factor model (FFM) of personality. To date, no studies have investigated the personality structure in relatives with the NEO-PI-R. In this study, there were three groups, 28 remitted bipolars, 21 1^st degree relatives of the bipolar group and 29 healthy controls. All subjects were administered the SCID-II, mood and anxiety modules, the Modified Hamilton Rating Scale for Depression (MHRSD), the Young Mania Rating Scale (YMRS) to assess remission and the NEO-PI-R. It was hypothesized that the relative group would show scores similar to the bipolar group and that both groups would be significantly different from the control group on Neuroticism, Extraversion and Conscientiousness with some predictions also made on the facets of these personality dimensions. The results revealed that the bipolar group was significantly different from both relatives and controls on Neuroticism, Openness and Conscientiousness, but not on all predicted facets and not on Extraversion. The most significant finding of this study was that the relative group appeared much more like the control group than the bipolar group. This suggests that the NEO-PI does not recognize differences in this sample of relatives. Possible conclusions to this result are that the five-factor personality model is not a recognizable endophenotype for the relative group. Another conclusion is that the personality structure of relatives serves as a “protective factor” from the illness.