The purpose of the study was to: (1) observe the frequency of giving way to pedestrians at the zebra crossing; (2) reveal speed adaptation problems; (3) examine if the time difference between the arrival of the pedestrian and the car have any influence on the speed of the approaching car; (4) identify so-called 'ideal' situations (in which the car brakes on the driver's own initiative in order to give way to a pedestrian) upon which recommendations can be given for implementing means to improve speed behaviour at zebra crossings. It was hypothesized that the speed behaviour of drivers approaching the zebra crossing depends on the pedestrian's arrival at the curb related to the car's expected arrival at the zebra crossing. The speed was measured on randomly selected 'free' passenger cars which approached a non-signalized mid-block zebra crossing on a two-lane arterial road. Every second a radar gun, hidden at the road side, sent the speed data to a lap-top computer in which the observer could also register pedestrians' arrival at, and start from the curb. Simultaneous video recordings were made in order to obtain a more detailed description of the interaction between the car and the pedestrian. Speed behaviour in encounters (148 observations), non-encounters with pedestrian presence (642 observations) and situations without pedestrian presence (690 observations) was compared. Situations with pedestrian priority were classified. The results show that the frequency of giving way is 5%. Drivers do not observe the law concerning speed behaviour at the zebra crossing, as they do not "adapt the speed in such way that they do not endanger pedestrians who are already on, or are about to step onto the zebra crossing". In encounters, three out of four drivers maintain the same speed or accelerate and only one out of four slows down or brakes. These results indicate that maintained high speed (even exceeding the speed limit of 50 km hour-1) is the signal from the drivers that they do not intend to give way to the pedestrian at the zebra crossing. The conclusion is that encounters between cars and pedestrians at the zebra crossing are critical situations in which the driver has to be influenced before he reaches the decision zone at 50 to 40 m before the zebra crossing in order to prevent the 'signalling by speed' behaviour. Countermeasures to improve driver behaviour at the zebra crossing are discussed.