Tenderness and pain thresholds in pericranial muscles were studied in a random sample of 735 adults aged 25-64 years. This study was a part of a multifaceted, epidemiological study of different headache disorders. Manual palpation and pressure pain threshold were performed by observers blinded to the persons' history of headache. The aim of the study was to evaluate the possible role of pericranial myofascial nociception in headache pathogenesis. Tenderness in migraineurs did not differ from non-migraineurs. Subjects with episodic tension-type headache and females with chronic tension-type headache were more tender than the rest of the population, and males without any experience of headache were less tender than the rest of the male population. A strong positive correlation between tenderness and frequency of tension-type headache was found (males: P < 10(-4); females: P < 10(-5)), while no relation between tenderness and migraine frequency was seen (P = 0.43). In subjects having actual headache at the day of examination tenderness was 32% increased compared to a matched group with identical usual frequency of headache, but without headache during the examination. A significant relation of tenderness to the recency of last episode of headache was detected in both sexes after control for usual frequency and actual headache (males: P < 10(-3); females: P < 10(-4)). Pressure pain thresholds were largely normal indicating normal pain processing and contradicting the idea that tension-type headache mainly is due to generally increased pain sensitivity. This study supports the pathogenetic importance of muscular factors in tension-type headache, while muscular factors are of no primary importance in migraine.