This paper describes a new interview questionnaire for the assessment of leisure-time physical activity (LTPA), and its validation as an alternative estimate of physical fitness. British subjects (77 males, 41 females) provided details of their LTPA over a period of a "typical" 2 weeks, enabling the estimation of the energy expended. Physical fitness was assessed with a battery of measures, the optimal single measure being sub-maximal physical work capacity (PWC). The questionnaire showed LTPA to be stable following test-retest administration (r = 0.86; p less than 0.0001) for total LTPA energy expenditure. Total LTPA was found to be significantly related to PWC (r = 0.48, p less than 0.0001), as were very hard (r = 0.55; p less than 0.0001) and hard LTPA (r = 0.38; p less than 0.0001). Multiple regression analysis with PWC as the dependent variable yielded a multiple correlation of r = 0.87, with significant contributions from very hard and hard LTPA. It is concluded that whilst this questionnaire is both reliable and a valid estimate of physical fitness amongst a population consistent in their leisure-time physical activities, there is scope for its further use within larger populations, allowing for an analysis of the effects of age and gender on the associations so far observed.