The serum DBH (Dopamine-β-Hydroxylase) activity in a total of 177 individuals was determined in this study during the period from June 1974 to June 1975. The subjects were 103 manic-depressive psychosis patients and 43 patients with other diseases who had visited the Neurology Department of the Hiroshima City Hospital and 31 healthy subjects included for comparison. Among the 103 manic-depressive psychosis cases, there were 55 cases, 40 with unipolar depression and 15 with bipolar depression, in whom determination of activity was made during both the depressive and remission phases. Further, in 25 unipolar depression and 6 bipolar depression cases, the diurnal variation of activity was measured and studied in detail. The following results were obtained from the above findings. 1. There was marked individual difference in serum DBH activity. Great individual difference was observed in healthy persons also, but in the same individual the value obtained by repeat test perform 9 months after the initial test showed almost the same result and the rate of variation was less than ± 10% . 2. Comparison of the mean values of serum DBH activity among various diseases showed the values to be high in the order of schizophrenia, neurosis, unipolar depression and bipolar depression, but the differences from healthy persons were not significant. 3. Variation in serum DBH activity in manic-depressive psychosis strongly suggested the heterogeneity of unipolar and bipolar depressions. Further it was also assumed that there were heterogeneous diseases mixed among the unipolar or bipolar depression cases, and thus further subclassification was performed. 4. Cases with unipolar depression were classified into 3 groups according to whether the differece in value of serum DBH activity was significantly higher in the depressive phase than the remission phase (Group Ⅰ), unchanged (Group Ⅱ) and significantly lowere (Group Ⅲ). The respective clinical characteristics were as follows: Group Ⅰ consisted of 40% of the unipolar depression cases, was made up predominantly of women and a strong feeling of anxiety prevailed, but the subjects responded well to tricyclic antidepressants. Group Ⅱ was composed of 32.5%, the subjects were highly irritant and many did not respond well to tricyclic antidepressants. Group Ⅲ consisted of 27.5%, was made up predominantly of males, severe psychomotor retardation was present and the subjects responded well to tricylic antidepressants. 5. All cases with bipolar depression presented the characteristic finding of lower activity values during the depressive phase than the remission phase. Cases with manic phase activity values lower than the remission phase were classified as Group Ⅰ and those with higher values Group Ⅱ. The respective clinical characteristics were as follows: Group Ⅰ was composed of 69.2% of the bipolar depression cases of whom many changed to mania within a short period of time after their depressive phase making it appear as if the depressive and manic phases were one single continuous phase. Group Ⅱ was composed of 30.8% and was frequently found in those with a marked manic phase only or in those whom the manic or depressive phase developed independently. 6. The diurnal variation of serum DBH activity in manic-depressive psychosis was not of a fixed pattern such as being low in the morning and high at noon. However, many presented a similar rhythm for both the depressive and remission phases of the disease, but the rate of diurnal variation was greater during the depressive or manic phase than the remission phase. The rate of diurnal variation during the remission phase was about ± 10%.