The average annual incidence of definite and probable MS in Gothenburg was re-investigated. For 1950-1954, 1955-1959 and 1960-1964 it was 4.2, 4.2 and 4.3/100,000/year. For the five-year periods between 1974 and 1988 it was 3.0, 2.7 and 2.0/100,000/year. If possible MS was included, the corresponding incidence for 1950-1964 was 5.2, 5.3 and 5.1, and for 1974-1988 it was 3.9, 3.9 and 4.3/100,000/year. Neurological methods and diagnostic criteria were constant throughout the period. The 1950-1964 incidence was based on personally investigated cases, while the 1974-1988 incidence was based partly on review of Gothenburg neurology records. It is concluded that there has been a significant decrease in the incidence of MS in this area. However, the notified decrease may partly be explained by alterations in the case ascertainment procedure. Since the Swedish measles vaccination program started in 1971, the occurrence of measles has been declining and has practically ceased during the 1980s. The time when a possible influence of mass vaccinations against childhood diseases on MS incidence can be monitored is discussed.