Serum testosterone does not correlate with androgen tissue activity, and it is critical to optimize tools to evaluate such activity in males. Ultrasound measurement of bulbocavernosus muscle (BCM) was used to assess the relationship between the number of CAG repeats (CAGn) in the androgen receptor (AR) and the BCM size; the changes in the number of CAGn over age were also evaluated. Transperineal ultrasound measurement of the BCM was also performed. AR CAGn were determined by high performance liquid chromatography, and morning hormone levels were determined using immunoassays. Forty-eight men had CAG repeat analysis. Twenty-five were <30 years of age, mean 23.7 years (s.d. = 3.24) and 23 were >45 years of age, mean 53 years (s.d. = 5.58). The median CAGn was 21 (13-29). BCM area was greater when the number of CAGn were <18 as compared to the number of CAGn >24 (P = 0.04). There was a linear correlation between the number of CAGn and the BCM area R 2 = 16% (P = 0.01). In the 45 to 65-years-old group, a much stronger negative correlation (R 2 = 29%, P = 0.01) was noticed. In the 19 to 29-years-old group, no such correlation was found (R 2 = 4%, P = 0.36). In older men, the number of CAGn increased with age (R 2 = 32%, P = 0.01). The number of CAGn in the AR correlates with the area of the BCM. Ultrasound assessment of the BCM is an effective surrogate to evaluate end-organ activity of androgens. The number of CAGn may increase with age.