Fluoridation of water is a controversial measure because of the suspicion that it has harmful effects on health. Opinions differ as to the reality of these fears. In Kuopio, after distressing disputes over the fluoridation issue, the City Council decided to stop fluoridation at the end of 1992. In fact, however, it was discontinued at the end of November, one month early, without the public being told. The aim of this study was to find out whether the occurrence of 25 selected symptoms was connected with exposure to fluoridated water. In order to do this we compared the prevalence of symptoms during the months before and after the undisclosed cessation of fluoridation and after the cessation had been officially announced. Postal inquiries concerning symptoms were sent to 1000 randomly selected adults in November, to a further 1000 in December 1992 and again to the same 2000 people in March 1993. The response rates were 40-26%. The percentage of those with two or more symptoms was the same (45%) in November and in December but decreased to 32% in March. The mean number of symptoms per respondent decreased from 1.9 in November to 1.4 in March (P < 0.001) and in December-March from 1.8 to 1.2. The decrease was most significant for symptoms related to the skin. Since the occurrence and mean number of symptoms were fairly similar during actual and supposed fluoridation, the results do not support the theory that the symptoms considered in this study are caused by the physical effect of fluoridated water. On the other hand, the significant reduction in the number of symptoms only after the respondents had become aware of the discontinuation of fluoridation reveals that fluoridation may have psychological effects which present as perceived symptoms.