The author discusses the validity of the borderline concept in terms of the recent empirical research. Five phases of exploration are reviewed; namely, clinical descriptions, laboratory studies, delineation from other disorders, follow-up studies, and family studies. These data are related to the current status of the diagnosis for clinical practice and to areas warranting further research. The conclusions reached show initial support for the validity of the borderline diagnosis. Descriptively, the literature tends to characterize these patients as having impulsivity, heightened affect, mild psychotic reactions and disturbed close relationships. Speculation about possible etiologies is felt to be premature and the identification of meaningful subgroups of patients within this syndrome is not yet possible. Further research using the objective scales in well-defined samples is felt to be required, as are family and prospective follow-up studies.