Changes in the incidence of testicular cancer in Victoria, Australia were studied from 1950 to 1985. Cases from the period 1950 to 1978 were derived from many sources (1116 cases). Emphasis was placed on diagnostic reliability; 97% of cases were confirmed histologically, and of these, 86% were reviewed at the Peter MacCallum Cancer Institute. For the years 1982 to 1985, Victorian Cancer Registry data were used. The incidence rose by a factor of 2.9 from 1.44 (95% confidence interval [CI], 1.15 to 1.73) per 100,000 in 1950 to 1954 to 4.16 (95% CI, 3.73 to 4.59) per 100,000 in 1982 to 1985. Between 1950 to 1954 and 1965 to 1969, there was a sharp rise, followed by a plateau or dip, then a further rise. Among seminomas, the rates rose in most adult age groups, whereas among nonseminomas, the rise was concentrated in younger age groups. There was a significant trend to earlier age of occurrence among nonseminomas (P = 0.0004) but not among seminomas (P = 0.89). Cohort analysis revealed a trend toward increasing rates for both seminomas and nonseminomas, and confirmed the trend toward earlier age of onset for nonseminomas. Disproportionate increases were observed for the 1915 to 1924 cohort of seminomas and the 1930 to 1939 cohort of nonseminomas. Analysis of available data from other Australian states indicated comparable rising incidence in New South Wales, Tasmania, and Western Australia.