In bipolar disorder, the factors provoking a new episode are unknown. As a seasonal variation has been noticed, it has been suggested that weather conditions may play a role. The aim of the study was to elucidate whether meteorological parameters influence the development of new bipolar phases. A group of patients with at least three previous hospitalizations for bipolar disorder was examined every 3 months for up to 3 years. At each examination an evaluation of the affective phase was made according to the Hamilton Depression Scale (HAM-D(17)), and the Bech-Rafaelsen Mania Rating Scale (MAS). In the same period, daily recordings from the Danish Meteorological Institute were received. We found no correlations between onset of bipolar episodes [defined as MAS score of 11 or more (mania) and as HAM-D(17) score of 12 or more (depression)] and any meteorological parameters. We found a statistical significant correlation between mean HAM-D(17) scores and change in mean and maximum temperature, and non-statistical significant correlations between mean MAS scores and rainfall plus atmospheric pressure, and non-statistical significant correlations between mean HAM-D(17) scores and hours of sunshine and cloudiness. Though meteorological factors may have an impact on triggering new episodes in bipolar patients, they do not constitute a dominant cause.