Genetic factors have long been implicated in the aetiology of bipolar disorder (BD). During the past two decades several linkage studies have been carried out with the aim of identifying major genes. However, remarkable discrepancies in results both between and within studies have constituted a major problem. In order to elucidate some of these conflicts, we assessed the published literature on linkage studies of bipolar disorder, focusing on methodological issues. Studies published between January 1980 and December 1994 were identified by computerized literature searches and subsequent scanning of review articles, and the reference lists of the articles primarily identified. A set of defined inclusion and exclusion criteria was used to select studies for assessment. A total of 31 variables were determined, and pre-defined codes were assigned in a structured manner. More than 200 citations were reviewed, and 60 articles were included in this study. Descriptive statistical analyses of the variables, as well as associations between variables, are presented. The findings are discussed with regard to the possibility that, beyond the genetic complexity of the disorder itself, there are several other similarly complicated study design issues which should be more carefully observed. Moreover, the need for standardization of basic criteria to use and report clinical and analytical parameters employed in linkage studies is strongly suggested.