The relationship of risk of testis cancer, and subdivisions of the tumour by histology, to variables which may be related to hormonal status, sexual behaviour and fertility was investigated in data from a case-control study with 259 cases and 2 sets of controls. No consistent association was found between testis cancer risk and age at puberty, need to shave, obesity, alcohol intake, animal fat intake, and sexual behaviour. There was a significant excess of seminomas in very tall men. Testis cancer cases showed lower fertility than controls according to various measures, but this reflected the greater frequency of cryptorchidism among cases compared to controls. Among non-cryptorchid subjects there was no clear evidence that infertility was associated with risk of testis cancer.