Three hundred thirty-two laborers in a shipyard, all of whom had been working for decades in heavy industry, were compared with regard to the occurrence of coxarthrosis with a similar sized group of white-collar workers and with a random population sample. All three sets of men were of the same age. About one-fourth (the same in all three groups) had been referred for a hip roentgenographic examination at some time over the previous decades. In the group of laborers and their white-collar controls, coxarthrosis occurred in about 3%, with no difference between the two samples. The occurrence was less, but not significantly less, in those men who had been randomly selected from the population.