This study was undertaken to measure and compare the thickness and cross-sectional area of the Achilles tendon between frequent- and infrequent-exercise subjects, and between the dominant and nondominant ankles in an asymptomatic Chinese population. Interobserver variability in the measurement of the size of Achilles tendon was also evaluated. High-resolution ultrasound (US) examination of Achilles tendons was performed in 40 healthy subjects (20 who frequently exercised, had exercise at least 3 days per week and at least 2 h per session; and 20 who infrequently exercised); their age range was 19 to 25 years. The thickness and cross-sectional area of the Achilles tendons were measured in a transverse scan at the level of medial malleolus. For each subject, the Achilles tendons were measured by five operators to evaluate the interobserver variability in the measurements. The mean thickness and cross-sectional areas of the Achilles tendon in a healthy Chinese population are 5.23 mm(2) and 56.91 mm(2)(2), respectively. The mean thickness of the Achilles tendon of frequent-exercise subjects (dominant ankle 5.43 mm, nondominant ankle 5.38 mm) was significantly greater than that of infrequent-exercise subjects (dominant ankle 5.08 mm, nondominant ankle 5.04 mm) (p < 0.05). The cross-sectional area of the tendons was also larger in frequent-exercise subjects but, whereas a significant result was found in dominant ankles (frequent-exercise subjects 60.46 mm(2)(2), infrequent-exercise subjects 54.71 mm(2)(2)) (p < 0.05), this was not the case for the nondominant ankles (frequent-exercise subjects 57.09 mm(2)(2), infrequent-exercise subjects 55.4 mm(2)(2)) (p > 0.05). In both frequent- and infrequent-exercise subjects, there was no significant difference in the mean thickness and cross-sectional area of Achilles tendon between dominant and nondominant ankles (p > 0.05). There was a high reproducibility in the sonographic measurement of the thickness (68%) and cross-sectional area (81%) of Achilles tendons. Results suggested that exercise would cause increase in the thickness and cross-sectional area of Achilles tendon. Interobserver variability is not significant in the sonographic measurement of Achilles tendons.