1. The factors influencing the seasonal variation in blood pressure measured at home in normotensive women were examined. 2. Sixteen female subjects (56.3 +/- 7.9 years old, mean +/- SD) measured their blood pressure and pulse rate at home each morning for more than 20 times per month for at least 1.5 years. Blood pressure and body weight were also determined in the office once or twice a month in that period. Monthly means of outdoor and indoor temperatures and daytime length were obtained from the Meteorological Observatory. The single cosinor method was used to evaluate circannual rhythm. 3. We observed a biphasic seasonal variation in self-recorded blood pressure measured at home, environmental temperature and daytime length but found no apparent seasonal variation in body weight and blood pressure measured in the office. The lowest levels of systolic and diastolic blood pressure measured at home were observed in July. The longest daytime length was recorded in June, while the highest outdoor temperature and indoor temperature were recorded in August, indicating that the longest daytime length preceded and the highest environmental temperature lagged behind the lowest level of blood pressure. The shortest daytime length is in December. The lowest outdoor and indoor temperature were observed in January, while the highest levels of self-recorded systolic blood pressure and diastolic blood pressure were in January and December, respectively. Half-amplitudes of self-recorded systolic and diastolic blood pressure were 2.6 +/- 1.0 mmHg and 2.0 +/- 0.8 mmHg, respectively. 4. These findings indicate the importance of a seasonal effect, i.e. daytime length and the environmental temperature, on the blood pressure of individuals.