162 depressed patients, consecutively admitted to the Department of Psychiatry, Umeå University, were rated by means of the comprehensive psychopathological rating scale (CPRS) within a few days after their admission, and before any antidepressive treatment had been implemented. The same patients were later subdivided into groups according to the presence or absence of psychiatric illness among their first-degree relatives. Four groups were thus formed: depression pure disease (DPD), i.e., comprised of patients with only secondary cases of depression; depression spectrum disease (DSD), comprised of patients with secondary cases of both depression and alcoholism and antisocial personality among their first-degree relatives; positive family history-others (FH+.O), included those patients with first-degree relatives affected by other psychiatric disorders than those required for inclusion among the DSD; and finally, sporadic depression (SD), comprised of those patients who had no family loading at all. The results showed that the severity of illness was similar in all four groups. Only a few symptomatological differences emerged, and these related consistently to a higher frequency of symptoms of retardation (inability to feel, lassitude, and concentration difficulties) in the DSD group as compared to the SD group. No significant symptomatological differences occurred between patients in the DPD and DSD groups. It is concluded that a classification of unipolar patients based on family data does not predict the occurrence of various symptomatological patterns.