The aim of this study was to investigate the interaction between the Volvo Lane Departure Warning System (LDW) and the road markings. LDW is based on a digital camera which basically registers what the driver’s eyes see. This implies that the road markings must be visible, i.e. there must be a certain level of the contrast between the road markings and the road surface to LDW to work. The visibility of the road markings in daylight conditions is quantified by the daylight luminance coefficient and in darkness, in head-light illumination, by the retroreflectivity. An essential question to answer was what level these two variables must meet to make the LDW to work. Is it good enough to fulfil the demands in the regulations? Another question to answer was to what extent the LDW gives undesired or incorrect alarms. Such malfunctions may cause the driver to switch off the system and should therefore not occur. The study has been accomplished in the field on public roads which were defined in advance. These roads were chosen so that several types of road markings - profiled as well as flat, new as well as worn - were tested. This fact meant that some of the road markings had good visibility in most light and weather conditions, while others were almost invisible, at least in darkness and in rain. The whole test road was 88 km long, with start and stop in Linköping. A Volvo S80 equipped with a LDW device was driven on the road in different weather and light conditions. During the journey, the accessibility, from which the frequency of missing alarms could be estimated, was registered. Furthermore, undesired and incorrect alarms were noted manually. A few days after the last test drive, the performance of the road markings on the test road was measured by using a mobile measurement system based on a reflectometer and an optocator. This system, the Road Marking Tester, can measure or predict the retroreflectivity and luminance coefficient of dry road markings and predict the retroreflectivity of wet road markings.