In this article, emphasis was placed on the relationship between exercise and upper respiratory tract infection (URTI) in humans, experimentally induced infections in animals subjected to varying levels of exertion, and potential changes in the immune system that might explain the altered risk of infection. With regard to induced infections in animals, the influence of any exercise intervention appears to be pathogen specific, and dependent on the species, age, and sex of the animals selected for study, and the type of exercise paradigm. In general, although further research with larger subject pools and improved study designs is needed, published data at this time support a "J" curve relationship between risk of URTI and increasing exercise workloads. For example, individuals exercising moderately may lower their risk of URTI while those undergoing heavy exercise regimens may have higher than normal risk. Although researchers have investigated changes in immune function that might provide a biological rationale for the "J" curve model of infection and exercise, the wide variety of research designs, exercise protocols, subject characteristics, and methodologies combined with the innate complexity of the immune system have made interpretation of published findings equivocal. T and NK cell function, for example, is often reported to be decreased during recovery from high-intensity exercise. However, when adjustments are made for exercise-induced perturbations in blood lymphocyte subsets, any link to decreased host protection is unlikely.