The cervical spine of young children is normally very mobile. Consequently,relatively little force is required to cause subluxation of the cervical vertebrae. In this respect, the incidence of cervical spine injuries in children involved in car accidents seems to be surprisingly low. However, recent European data seem to indicate that serious cervical spine injuries occur more frequently than had previously been thought. In particular, small children seated in forward facing restraint systems with a harness belt appear to be at risk from tetraplegia in moderate frontal impacts.To investigate the significance of these findings, TNO has undertaken a research programme to measure the neck loads induced in TNO-P3/4 child dummies seated in various types of restraint system. A series of dynamic sled tests has been performed in which the neck loads induced in a TNO dummy have been measured using a special transducer, which replaces the Atlas-Axis block of the dummy. The design of this transducer and the test results that have been obtained to date are discussed in detail in the present paper. Special attention is focused on how the misuse of restraint systems can affect neck loads, and on the influence of different seating positions. In addition, consideration is given to the significance of free flexural/extensional motions of the P3/4 head. Finally, forward facing and rearward facing systems are compared with each other and child dummy neck loads contrasted with proposed adult tolerance limits.In parallel with the experimental work, a series of mathematical simulations has been performed, using the MADYMO CVS programme. These analyses have been aimed at studying the effect of chin-to-chest contacts and assessing the impact of headlneck joint characteristics on neck loads. A “Design of Experiments” programme allowing systematic parameter variations has been used to analyse the significance of various test and design parameters.THE SAFETY OF CHILDREN in passenger cars is an important area in the field of passive safety. Parents, product designers, legislators and researchers have a special duty to this vulnerable group of road users, since young children are not able to make judgements about safety themselves. Severe injuries not only have implications for children and their families in the longer term, but they also place a considerable burden on society.In the last 10 years, there have been substantial improvements in the design of child restraint systems. Accident investigations in Europe have shown that there are significant advantages in using child restraint systems. Langwieder and Hummel (*), for instance, found that for unrestrained children, the risk of serious/fatal injuries is seven times higher than for restrained children. However, their accident data also indicated the occurrence of serious neck injuries in young children restrained by forward facing systems.At a workshop on child safety in 1986 , attended by some 100 experts in this field, serious neck injuries were not considered to be important since they were rarely observed in the accident data available at that time. This view was given further credence by Carlsson et al. , who concluded that serious neck injuries were very rare and that in the case of restrained children, only 0.9% were within the AIS 2-6 range.Langwieder and Hummel  reported that neck injuries rank second to head injuries for restrained children. Of all the injuries reported for restrained children, 14% appear to be AIS class 1 neck injuries and 1% AIS class 5. No serious neck injuries were found in comparable data for unrestrained children. Using these results as a basis, Langwieder and Hummel carried out further research in Germany to investigate the extent to which children using restraint systems sustained serious neck injuries in car accidents. In a recent paper , they reported on 5 cases of tetraplegia. Four of the children injured in this way were restrained in systems with a 4-point harness belt, when the cars in which they were travelling became involved in ‘moderately severe’ head-on collisions. News of these cases led to the advantages and disadvantages of forward facing versus rear facing child restraint systems being discussed in Europe. It was felt that such cases could not be ignored even though the frequency of serious neck injuries in children appeared to be very low. Moreover, there was agreement that more attention should be given to studying these incidents as this could ultimately lead to better products and improved legislation.