There has recently been a substantial change in our concept of periodontal disease and particular attention is now focused on that small proportion of the population who appear susceptible to its more aggressive forms rather than the majority in whom bone loss progresses very slowly. It is also apparent that the presently available clinical parameters are of little value in predicting future destructive activity. Under these circumstances, the aim of this paper is to review the contribution of radiography to the diagnosis of chronic periodontitis as traditionally perceived and then reassess its status in the light of these newer concepts. Panoramic radiography, followed by the appropriate periapical radiographs (taken with the paralleling technique), is proposed as an alternative to complete mouth intra-oral surveys on grounds of both diagnostic yield and radiation thrift. The five areas to which radiography, despite its limitations, can make a significant contribution are in the assessment of bone loss, mobility, occlusal trauma, calculus and marginal overhangs and crown-root ratio. The validity of the three criteria that have been proposed for the radiographic assessment of early periodontitis, loss of crestal bone height, marginal widening of the periodontal ligament and crestal irregularity, is evaluated in detail and it is concluded that only the first is of any diagnostic worth, providing at least two sequential radiographs are available. While there is an urgent need to develop techniques of greater sensitivity for the early identification of periodontal bone loss, there must be some doubt as to the value of any bone imaging technique in predicting the susceptible patient. Follow-up radiography should be limited to these sites showing clinical evidence of further disease activity.