Serologic surveys, undoubtedly the most accurate of the various methods of determining the prevalence of syphilis, are limited in precision by certain factors, such as the size and randomness of the sample, the specificity and sensitivity of the tests used and the failure to detect seronegative syphilis. During the course of epidemiologic studies of leprosy in 1939 and 1940, more than 70 per cent of the inhabitants of the island of St. Thomas and more than 95 per cent of the population of a rural area on the island of St. Croix were examined in clinics. Serologic tests were made on nearly 10,000 persons, representing 60 per cent of the inhabitants of the two areas. This number is felt to be a true random sample, and, furthermore, the serologic tests were considered to be sufficiently accurate and sensitive. Seronegative syphilis remains an unknown quantity. It is the object of this