This study compared 3 d of carbohydrate loading (CHOL; 8.4 g x kg(-1) x d(-1) carbohydrate) in female eumenorrheic athletes with 3 d of an isoenergetic normal diet (NORM; 5.2 g x kg(-1) x d(-1) carbohydrate) and examined the effect of menstrual-cycle phase on performance, muscle-glycogen concentration [glyc], and substrate utilization. Nine moderately trained eumenorrheic women cycled in an intermittent protocol varying in intensity from 45% to 75% VO2max for 75 min, followed by a 16-km time trial at the midfollicular (MF) and midluteal (ML) phases of the menstrual cycle on NORM and CHOL. Time-trial performance was not affected by diet (CHOL 26.10 +/- 1.04 min, NORM 26.16 +/- 1.35 min; P = 0.494) or menstrual-cycle phase (MF 26.05 +/- 1.10 min, ML 26.23 +/- 1.33 min; P = 0.370). Resting [glyc] was lowest in the MF phase after NORM (575 +/- 145 mmol x kg(-1) x dw(-1)), compared with the MF phase after CHOL (728 mmol x kg(-1) x dw(-1)) and the ML phase after CHOL and NORM (756 and 771 mmol x kg(-1) x dw(-1), respectively). No effect of phase on substrate utilization during exercise was observed. These data support previous observations of greater resting [glyc] in the ML than the MF phase of the menstrual cycle and suggest that lower glycogen storage in the MF phase can be overcome by carbohydrate loading.