The effectiveness of long term lithium prophylaxis has been studied in thirty patients with bipolar affective disorder. The study covered only these patients receiving prophylactic lithium treatment for at least 10 years, who demonstrated positive response to lithium during the initial five years of the therapy. Standard medical documentation was analyzed for the time-course of the illness within the whole observation period (number, duration and character of remissions) as well as for the history of lithium therapy (serum lithium concentrations, regularity of medical checks, breaks in the lithium treatment). The effectiveness of lithium prophylaxis has been found to diminish with time in as many as 40% of patients subjected to the prolonged lithium therapy, as evidenced by an increased number of recurrences in this group. The explanation for this phenomenon is not clear, particularly that the change in the responsiveness to lithium has been found not to correlate with age, duration of illness, duration of lithium therapy, incidence of somatic diseases and history of lithium treatment (serum lithium concentrations, breaks in the treatment).