The paper describes a study of pedal cycle accidents focussing on helmet effectiveness. Only accidents in which a helmet was worn and received an impact were studied. Cases were collected from hospital accident and emergency units, through the police and via direct contact. Forty-two cases, all helmeted were examined. There were four fatal accidents; all four involved a collision with a motor vehicle. Nine cases experienced a head injury of AIS severity > or = 2, although there were no skull fractures. Helmet impacts tended to be close to the rim anterio-laterally. The majority of non-fatal (AIS > or = 2) head injury cases received a helmet impact to the anterio-lateral rim, which corresponds to the temporal/parietal region of the head. This site received directly only 25% of the impacts, and of these impacts, 75% produced head injuries of at least AIS = 2. Soft-shell helmets tended to disintegrate on impact, and although only a single impact occurred, a helmet should remain intact to provide protection during second impacts. There was a general increase in the percentage of subjects injured or killed in accidents that involved a second vehicle compared to single vehicle accidents, 54% to 44%, respectively. This trend was stronger with cars travelling at greater than 30 km/h.