Using direct surveys in the surveillance of occupational illness and injury was discussed. The features of direct surveys were summarized. A direct survey can be defined as a procedure in which information is collected systematically from subjects that had been selected through a specific sampling procedure to be representative of larger populations. Direct surveys were considered to be effective methods for obtaining surveillance data for many job related conditions or diseases. They were especially useful for identifying conditions that cause considerable morbidity, but little or no mortality, such as noise induced hearing loss or occupational dermatitis, which may not be easily recognized as being work related. Examples of national direct surveys were given. These include the National Occupational Hazard Survey, the National Study of Coal Workers Pneumoconiosis, the National Occupational Exposure Survey, and the National Occupational Health Survey of Mining conducted by NIOSH and the National Health Interview Survey, the National Natality Survey, the National Maternal and Infant Health Survey, and the National Health and Nutrition Examination surveys conducted by the National Center for Health Statistics (NCHS). The role of periodic surveys in surveillance was discussed. Periodic surveys, especially the large national surveys, present opportunities to standardize methods to be used in smaller surveillance surveys. They can also provide normative or reference databases for various parameters of health status or adverse health effects. Specific future plans for NIOSH and the NCHS were described. The author concludes that direct surveys can obtain health information on persons that cannot be collected in other ways. They are especially useful for identifying conditions that cause considerable morbidity, but little mortality.