In a collision between a car and pedestrian, the location and severity of impacts to the pedestrian vary considerably with respect to the shape of the car front. Consequently for the three primary areas of a car that are important for pedestrian safety (the bumper, bonnet leading edge, and bonnet top) the protection requirements will also vary considerably acccordingt o the car shape. Any sub-system tests to assess the protection afforded by a particular design of car must take account of the important interactions between the vehicle structure and the corresponding body regions that are liable to suffer serious injury in an impact. In order to obtain further data to evaluflte these interactions, the Transport and Road Research Laboratory has conducted two series of full scale tests with a range of simulated car shapes. Impact velocities. force and acceleration were measured at each of the three primary impact points and these have been used to derive effective mass and energy at each phase of an impact. This paper summarises results of tests at 40 km/h using dummies representing a six year old child, and an adult pedestrian. The results show that bonnet height strongly influences impact energy for impacts to the bonnet leading edge and bonnet top. Bumper height influences the impact energy for bumper and bonnet leading edge contacts. The variations in impact severity are discussed with respect to their implications for sub-systems testing. The results in this paper can, for any style of car, identify structures where pedestrian protection requirements may be particularly demanding and also give guidance on the amount of protection required.