We present a comparison of data from two morbidity surveys conducted in 1970-71--the Second National Morbidity Study (subsequently referred to as NMS) and the General Household Survey (GHS). In the NMS 115 general practitioners recorded for one year details of the utilization of services by and the morbidity in their patients; and in the GHS the same sort of information was obtained from interviewing members of approximately 15,000 households. Differences in consultation and home visiting rates are examined. For this purpose, data in the NMS which were based on the verified population (population linked to the National Census for 1971) are contrasted with findings from the GHS (with telephone contacts excluded). Apart from small differences in home visiting rates especially among old people, the overall consultation rates are extremely close. We suggest that the losses to both studies, which occur especially among young mobile people in urban populations, need to be considered carefully when interpreting the data. Nevertheless, the doctor-generated NMS data matches well with the patient-generated GHS. Ways are suggested whereby the surveys might be more closely linked.