The purpose of this study was to determine the effects of 6 months of moderate aerobic exercise on age-dysregulated measures of T lymphocyte and natural killer (NK) cell number and function. Previously sedentary elderly (age = 65 +/- 0.8 years) subjects were randomly assigned to supervised 3 time/week exercise intervention group (EXC, n = 14) or flexibility/toning control group (FT-CON, n = 15). Fasting resting blood samples were drawn prior to and after the 6 month intervention. The EXC group exhibited a significant (P < 0.05) 20% increase in VO2 max, whereas the FT-CON group had a smaller non-significant (P = 0.07) increase (9%). Immune results revealed that, in general, changes in immune function in response to 6 months of exercise training at an average intensity of 52% heart rate reserve (HRR) were similar when compared to FT-CON who exercised at approximately 21% HRR. There were no intervention-induced changes in total white blood cell, neutrophil, lymphocyte, monocyte, eosinophil, or basophil blood counts. Furthermore, the percentage and number of CD3+, CD4+ and CD8+ T cells in the blood remained unchanged. There was a tendency for the percentage and number of CD4+ and CD8+ näive cells (CD45RA+) to increase and for CD4+ memory cells (CD45RO+) to decrease post-intervention, especially in FT-CON. Both groups exhibited a small intervention-induced increase in the T-cell proliferative response to mitogenic stimulation: the percentage change of which was higher in the EXC group at several doses of Con A. Unstimulated NK cell cytolysis versus K562 cells tended to increase (P < 0.1) in the EXC group with little change in FT-CON. We conclude that 6 months of supervised exercise training can lead to nominal increases in some measures of immune function, while not affecting others, in previously sedentary elderly.