The therapeutic index of docetaxel, doxorubicin and their combination may be improved by an adequate selection of the circadian time of administration. The present study constitutes a prerequisite to testing the clinical relevance of chronotherapy in human breast cancer. Three experiments were performed in C3H/HeN mice. Each treatment modality was administered i.v. once a week for 3 weeks at one of six circadian stages, during the light span, when the mice were resting: 3, 7, and 11 h after light onset (HALO), or during darkness, when the mice were active: 15, 19, and 23 HALO. The circadian time dependency of single agent tolerability was investigated in healthy mice using four dose levels for docetaxel (38.8, 23.3, 14, and 8.4 mg/kg/injection) and for doxorubicin (13.8, 8.3, 5 and 3 mg/kg/injection; experiment 1). The circadian time dependency of each single agent efficacy was studied in MA13/C-bearing mice, using two dose levels of docetaxel (38.8 or 23.3 mg/kg/injection) or doxorubicin (8.3 or 5 mg/kg/injection; experiment 2). The toxicity and the efficacy of the simultaneous docetaxel-doxorubicin combination were assessed as a function of dosing time in experiment 3. Two combinations were tested (A, 16.3 mg/kg/injection of docetaxel and 2.5 mg/kg/injection of doxorubicin; and B, 11.6 and 3.5 mg/kg/injection, respectively) at each of the above six circadian times. Mortality, body weight change, and tumor size were recorded for 60–70 days in each experiment. Single agent docetaxel or doxorubicin was significantly best tolerated near the middle of the rest span (7 HALO) and most toxic in the middle of the activity phase (19 HALO). Docetaxel or doxorubicin as a single drug were also most effective at 7 HALO, irrespective of dose. Treatment at 7 HALO produced highest rates of complete tumor inhibition (81% versus 11% at 3 HALO for docetaxel, p from χ² <0.001, and 69% versus 44% at 11 HALO for doxorubicin, not significant) and highest day 60 survival rate (100% versus 28% at 3 HALO for docetaxel, p from χ² <0.001 and 89% versus 69% at 15 HALO for doxorubicin, not significant). Docetaxel-doxorubicin combinations were most effective following dosing in the beginning of the rest span or short after the onset of the activity span, with regard to the rates of both complete tumor inhibitions (45% at 3 HALO versus 15% at 19 HALO) and day 70 survival rates (85% and 80% at 3 and 7 HALO respectively, versus 20% at 19 HALO). The efficacy of single agent docetaxel or doxorubicin and that of their combination varied largely as a function of circadian dosing time. Single agent docetaxel at 7 HALO was the best treatment option in this model with regard to both tolerability and efficacy. This timing may correspond to the middle of the night in cancer patients.