OBJECTIVE: To systematically investigate the effects of various shoe features (sole hardness, heel height, heel collar height, tread pattern) on dynamic balance control and perceptions of comfort and stability in young and older people walking over even and uneven surfaces. DESIGN: A mixed-design 3-way repeated measures with age as a between-subjects factor and surface and shoe conditions as within-subjects factors. SETTING: Gait laboratory. PARTICIPANTS: Young adults (n=11) and community-dwelling older adults (n=15). INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURES: Center of mass (COM)-base of support (BOS) margins, vertical and braking loading rates, and subjective ratings of perceived shoe comfort and stability. RESULTS: Overall, compared with the standard shoes, the soft sole shoes led to greater lateral COM-BOS margin (P<.001), whereas the elevated heel shoes caused reductions in posterior COM-BOS margin (P=.001) and in vertical and braking loading rates (both, P<.001). Subjects rated the elevated heel shoes as significantly less comfortable (P<.001) and less stable (P<.001) than the standard shoes. Only the young subjects perceived the soft-sole shoes to be less stable than the standard shoes (P=.003). CONCLUSIONS: Both young and older subjects adopted a conservative walking pattern in the elevated heel shoes and exhibited impaired mediolateral balance control in the soft-sole shoes. In contrast, increased sole hardness (above that found in a standard shoe), a tread sole, and a raised collar height did not improve walking stability in either group. It is concluded that shoes with elevated heels or soft soles should not be recommended for older people and that a standard laced shoe with a low collar and a sole of standard hardness with or without a tread provides optimal dynamic stability when walking on even and uneven surfaces.