Previous data on the effects of a single dose of hydroxyurea on C3H mouse duodenum and mammary tumors from a fast growing line (S102F) were used to predict times that may be optimal (i.e., minimize killing of the duodenal S-phase cells while enhancing the killing of tumor S-phase cells) for the administration of subsequent doses of hydroxyurea. These predicted protocols were tested by giving tumor-bearing mice injections of 2 doses at 24 hr intervals. A preliminary in vivo tumor treatment experiment was also done wherein multiple doses (up to 10) were given either at 12, 20, or 24 hr intervals with the mouse survival, body weights, and tumor volumes being recorded daily. The data show that partial cell synchronization was achieved in both tissues and the initial knetics of the surviving cells was essentially the same after a single dose, 2 doses, or 4 doses of hydoxyurea. Also, the different intervals between the 2 doses did not affect the timing of the initital peaks of DNA synthesis in partially synchronized cells; however, the height of the peaks was affected The results demonstrate that kinetic data can be useful for predicting optimal intervals for 2-dose regimes and probably multiple-dose regimes involving a single cell-cycle phase-speeific drug when applied to a mouse tumor model. However, the recovery phenomena in the respective tissues are extremely complicated and more animal tumor data need to be collected before one can make adquate use of cell-synchronizing agents and perturbed cellular kinetic data for routine clinical chemotherapy or combined modality therapy.