An account is given of the development of a multiphasic screening unit designed to detect susceptibility to and early evidence of cardiorespiratory disease in apparently healthy populations. The procedure is based on an augmented mass miniature radiography procedure costing £2.65 for each person examined. Some preliminary results are presented from the first prospective study of 4, 681 subjects re-examined after more than two years. The prevalence of cardiorespiratory abnormality and 271 deaths representing annual mortality rates of 10.0 per thousand in males and 3.8 per thousand in females respectively, are described. A comparison is made of certain measurements obtained at the first screening examination of male decedents and a random sample of survivors; and the use of the likelihood ratio, a statistical procedure derived from multivariate analysis, is suggested as a prognostic index of incipient cardiorespiratory disease worthy of further study. At present theoretical errors of up to 10 per cent exist in predicting survival of men under fifty and of up to 12 per centin predicting death for men over fifty. Larger errors may be present in the prediction of death in men under fifty and survival of men over fifty. The implication of these errors is discussed in relation to their consequence for the patient and their place in the validation of future surveys.