Two hundred and ninety-seven consecutive Charnley total hip replacements that had been followed for at least twenty years or until revision or death were analyzed to determine the effect of early debonding of the smooth-surfaced femoral component on its subsequent survival. Radiographically evident debonding was not found to have a significant effect, with the numbers available, on the long-term survival of the femoral component when the maximum thickness of the radiolucent line between the superolateral border of the prosthesis and the cement had been less than 2.0 millimeters during the first one to five years after the operation. The radiographic finding of debonding also was not found to be associated with pain in the hip. These data show that most components with early debonding functioned well during a long period of follow-up and suggest that debonding of a smooth femoral component of a Charnley total hip replacement should not be considered to be analogous to loosening. In contrast, when the maximum thickness of the radiolucent line between the superolateral border of the prosthesis and the cement was 2.0 millimeters or more, an early appearance of debonding was associated with a significantly poorer (p < 0.0001) probability of survival of the Charnley femoral component without revision because of aseptic loosening. Thus, pronounced early subsidence of the component within the cement mantle had a strong negative impact on the long-term performance of the implant. The results of the present study should not be extrapolated to prostheses with substantially different design characteristics, as it appears that different types of femoral components behave differently when debonding occurs.