Forty healthy males (M) and females (F) divided into two different age groups i.e. M50 years (range 44–57; n= 9), F50 years (range 43–54; n= 9), M70 years (range 64–73; n= 11) and F70 years (range 63–73; n= 11) volunteered as subjects for examination of muscle cross-sectional area (CSA) and maximal voluntary isometric force production characteristics of the leg extensor muscles and serum androgen and sex hormone binding globulin (SHBG) concentrations. The CSA in the male groups was greatly larger (P < 0.01) than in the female groups and both elderly groups demonstrated slightly (n.s.) smaller values in the CSA than the two middle-aged groups. Maximal force of 2854 ± 452 N in M50 was greater (P < 0.05) than that of 2627 ± 752 N recorded for F50 as well as the force of 2787 ± 843 in M70 was greater (P < 0.001) than that of 1849 ± 295 recorded for F70. The force between F50 and F70 differed significantly (P < 0.05) from each other. The maximal rate of force production in M50 was greater (P < 0.01) than in F50 as well as in M70 greater (P < 0.001) than in F70. Both middle-aged groups demonstrated greater (P < 0.05) values than the respective elderly groups of the same sex. The individual values in the CSA correlated with the values in maximal force both in the middle-aged subjects (r= 0.66; P < 0.01) and in the elderly subjects (r= 0.69; P < 0.01). The mean concentration of serum testosterone in M50 was slightly (n.s.) greater than in M70 and in F50 significantly (P < 0.05) greater than in F70. Serum SHBG levels were lower in the males (P < 0.01) than in the females and serum testosterone/SHBG ratio in M70 and in F70 were lower (P < 0.05) than in M50 and in F50, respectively. In the females significant positive correlations were observed between the individual values in serum testosterone concentration and the values both in the CSA (r= 0.46; P < 0.05) and in maximal force (r= 0.62; P < 0.01) as well as between serum testosterone/SHBG ratio and both the CSA (r= 0.55; P < 0.05) and maximal force (r= 0.68; P < 0.01). The present results imply that the decreasing basal level of blood testosterone over the years in aging people, especially in females, may lead to decreasing anabolic effects on muscles thus having an association with age-related declines in the maximal voluntary neuromuscular performance capacity in aging people.