The purpose of the present study was to investigate whether the muscle fibre composition is related to the number of muscle fibres. To resolve this issue, we developed fast-twitch fibre dominant rats (FFDR) by selective breeding and compared the findings to those of control rats (CR) obtained by random breeding. Percentage of type I fibres of the deep portion of gastrocnemius (DG), soleus (SOL), vastus intermedius (VI), adductor longus (AL), and biceps brachii (BB) muscles in FFDR were lower than CR. Percentage of type IIB fibres in DG, VI and AL and percentage of type IIA fibres of SOL in FFDR were higher than CR. However, fibre composition of plantaris (PLAN), extensor digitorum longus (EDL), rectus abdominis (RA), diaphragm (DIA), and palmaris longus (PL) muscles in FFDR were identical with CR. Total fibre numbers on the cross-sectional area in SOL, PLAN, EDL, AL and PL were counted. Numbers of type I fibres of all those muscles in FFDR were not different from CR. Numbers of type IIA fibres of SOL and AL and of type IIB fibres of AL in FFDR were greater than CR, but there were no significant differences in the number of type IIA or type IIB fibres of PLAN, EDL or PL between the two groups. Based on these observations, it is suggested that there are pleiotropic and muscle-specific effects on muscle fibre composition. In addition, the number of type II fibres is a possible determinant of muscle fibre composition.