The effects of repeated bouts of submaximal cycle ergometry exercise on changes in the percentage of perpipheral blood T-lymphocytes, the T-helper/inducer and T-cytotoxic/suppressor subsets, and natural killer (NK) cells were studied in 18 healthy young men who had no history of regular exercise training. Subjects were matched on the basis of maximal O2 uptake and assigned randomly to exercise or control groups, with controls resting quietly during the exercise sessions. The percentage of peripheral blood mononuclear leukocytes that reacted with monoclonal antibodies specific for T-lymphocytes (CD3+ cells), the helper/inducer subset (CD4+ cells) and cytotoxic/suppressor subset (CD8+ cells) of T-lymphocytes, and cells with NK activity (Leu7+ cells) were enumerated by fluorescence-activated flow cytometry for samples obtained immediately before and after exercise on days 1, 3, and 5 of a 5-day exercise regimen. The results of this study were mixed with decreases in the percentage of T-lymphocytes before vs. after exercise on days 1 and 3 (P < 0.001), a decrease in the percentage of T-helper/inducer cells before vs. after exercise on day 3 (P < 0.05), no effect of exercise on the percentage of T-cytotoxic/suppressor cells, and a marked increase in the percentage of NK cells after exercise on days 1 (P < 0.05) and 3 (P < 0.01). The total number of recovered NK cells in the mononuclear leukocyte fraction of blood also increased significantly after exercise on days 1 (P < 0.05) and 3 (P < 0.01). These findings 1) suggest that repeated exposure to submaximal exercise results in consistent increases in the percentage of NK cells, 2) demonstrate that the exercise effect on T-lymphocyte subset percents were variable over time, 3) confirm earlier reports on the impact of a single bout of submaximal exercise on percent and numerical shifts in peripheral blood T-lymphocytes and NK cells, and 4) extend the ings of earlier studies to include individuals of low fitness levels.