Executive dysfunctions are considered to be putative markers of familial/genetic vulnerability to both schizophrenia and bipolar disorder. However, familial resemblance must be demonstrated before executive functions are used as a potential endophenotype. The aim of this study was to investigate familial resemblance for executive functions in families of schizophrenic and bipolar subjects. We assessed executive functions by means of two tests - the Wisconsin Card Sorting Test (WCST) and the Trail Making Test (TMT) - in 351 subjects from five populations: schizophrenic patients, bipolar patients, a group of relatives for each patient group and controls. For both tests, cognitive assessment results were consistent with previous studies: schizophrenic patients showed the greatest impairment, followed by bipolar patients and then the two groups of relatives. In families of bipolar patients we observed familial resemblance for the WCST and part A and part B of the TMT. However, by contrast with the classical point of view, considering executive measures to be markers of genetic vulnerability to schizophrenia, we did not demonstrate familial resemblance for either of the two executive tests in families of schizophrenic patients. Thus, executive measures, as assessed by the WCST or the TMT, should not be used as endophenotypes in genetic studies of schizophrenia unless confounders are identified and their effects eliminated.