The benefits of physical activity in children are well documented. However, the relationship between physical activity, motor fitness and self-concept is not clear. Whereas some research affirmed the relationship between physical activity with motor fitness (e.g. Okely, Booth & Patterson, 2001, Raudespp & Jürimäe, 1996) and self-concept (e.g. Greene & Ignico, 1995), contradictory evidence refuting its relationship with motor fitness (e.g. Oja & Jürimäe, 2003) and self-concept (e.g. Tardie, 2000) were also available. The aim of this study was to clarify the relationship between physical activity, motor fitness and physical self-perceptions in Hong Kong school children aged 9-12. Five hundred and forty-five children (285 boys, 260 girls) in grades 4-6 from 7 primary schools completed the Physical Activity Questionnaire for Older Children (Kowalski, Crocker, & Faulkner, 1997) and the 22-items Chinese-version of Physical Self-Perceptions for Children (C-PSPP-C) adapted from Fox and Corbin’s (1989) scale and validated by Chow (2004) in a previous study on Hong Kong school children. Children’s gross motor ability was assessed by the Body Coordination Test developed in Germany by Schiling (1974) which yielded a Motor Quotient score. Effects of gender and age on physical activity, motor quotient and physical self-perception subscale scores were analyzed by means of 2 X 4 ANOVAs. Correlations among those three dependent variables were also obtained. Results suggested that boys were significantly more physically active (F=10.51, p<.05) and had higher motor ability (F=14.12, p<.05) than girls. A downward trend in physical activity was observed for girls with age 10 being the turning point. This pattern was not evident in boys. Younger children (ages 9-11) had higher motor ability than older children. Furthermore, boys had higher ratings in self-perceptions on all subscales in C-PSPP-C, except the physical condition subscale. Significant relationships were found between physical activity and motor fitness in both boys (r=.17) and girls (r=.17). When boys’ physical activity was correlated with self-perceptions, significant relationships (r =.26 to .39) were found for all subscales with the exception of physical strength. In girls, significant relationships (r=.16 to .36) were found between physical activity and all subscales of self-perceptions. To conclude, children's motor fitness and physical self-perceptions are significant correlates of physical activity.Keyword(s): physical activity