Basic single particle impact studies were carried out on copper, steel and dural. By calculating the coefficient of restitution without employing the usual small indentation approximations, closer agreement was found with experimental data. The experimentally determined dynamic hardness was found to be between 1.3 and 1.4 times the quasi-static indentation hardness. A notable feature of these experiments was the formation of surface rings in the bottom of the impact crater. These impacts were analysed in a manner analogous to that used for the impact of a liquid drop on a rigid substrate where damage is caused by the detachment of the shock wave from the contact periphery. An extension of the single particle impact studies was an investigation into the chipping car paint. The failure mechanism was found to be primarily shear debonding at the coating/substrate interface which facilitated the formation of chips and subsequent corrosion of the metal substrate. Suggestions were made to improve the performance of the paint system. In the area of multiple particle impact, considerable improvements were made to the existing erosion apparatus and associated instrumentation, particularly in the area of particle velocity measurement. This was then used to investigate the behaviour of a range of materials including pure metals, nimonic alloys and polymer specimens. The nimonic alloys were found to have an angular response more typical of a brittle material than of a ductile metal. Most of the materials studied exhibited incubation times which were shown to correspond to between ten and several hundred impacts on the same target area. Study of titanium metal (which emits sparks during erosion) at normal incidence showed that some mass loss is actually occurring throughout the incubation period, although the specimen may be exhibiting a net mass gain. This mass loss is thought to be due to a single impact mechanism such as cutting or ploughing, which is then swamped by a co-operative mechanism to produce a linear erosion rate. The incubation period cannot therefore be due solely to the superposition of a linear mass loss and a saturating rate of embeddment.