Low back disorders and trunk muscle function were studied by questionnaire, interview, clinical examination, and performance tests in 902 employees in the metal industry. The survey was repeated after a 10-yr interval for 654 persons. For those who were allowed to participate in the muscle function tests, the prevalences of clinical findings in the low back and of chronic lumbosacral disease were inversely associated with muscle function in both cross-sectional studies. This association seems not to be explained by the interference of pain with muscle function tests. There was no association between muscle function at baseline and either the 10-yr incidence of chronic low back disease or the development of low back symptoms. Men with poor muscle function at baseline had a slightly elevated risk for the appearance of clinical findings at follow-up. There was an inverse association between the clinical findings at baseline and muscle function at follow-up. The association between low back disorder and trunk muscle function seems to result from an effect of the former on the latter, but poor muscle function may also contribute to the risk of disorder.