The purpose of this study was to examine the relationships between dual-energy X-ray absorptiometry (DXA)-determined appendicular lean mass (aLM) and ultrasound-measured thigh muscle thickness (MTH) ratio and between aLM or thigh MTH ratio and zigzag walking performance. Eighty-one middle-aged and older adults (41 men and 40 women) aged 50 to 74 years volunteered for the study. Approximately two thirds of the subjects (34 men and 17 women) carried out regular sports activity (at least >2 times a week) including running and cycling exercise. MTH was measured using B-mode ultrasound at two sites on the anterior (A50) and posterior (P50) aspects of the mid-thigh. A50:P50 MTH ratio was calculated to evaluate site-specific thigh muscle loss. aLM and percent body fat were also determined using a DXA. Men had lower body fat and higher aLM than women. Anterior and posterior thigh MTH as well as A50:P50 MTH ratio was higher in men than in women. Zigzag walking time was faster in men than in women. Anterior and posterior thigh MTH was positively (p < 0.001) correlated to aLM and aLM index in men and women. However, A50:P50 MTH ratio was not significantly correlated with aLM and aLM index in both sexes. There was no significant correlation between aLM index and zigzag walking time in men and women. A50:P50 MTH ratio was inversely (p < 0.05) correlated to zigzag walking time in both men and women. Our results suggest that thigh MTH ratio is independent of age-related muscle mass loss detected by aLM.