Local recurrence occurs in 4–47% of breast cancer patients and is often associated with development of metastatic foci and resistant cell populations. Thus, recurrent breast cancer indicates a poor prognosis for the patient. Local tumor-derived 13762NF rat mammary adenocarcinoma cell clone MTF7(T20) was injected into the inguinal mammary fat pad and allowed to grow before surgical excision. Individual locally growing (primary) tumors were removed and established in short-term tissue culture. Corresponding local recurrences were excised after regrowth and established in short-term tissue culture. All sublines were tested for in vitro sensitivities to 5-fluoro-2′-deoxyuridine, Adriamycin, and ionizing X-irradiation. Using a clonogenic colony formation assay, responses of individual sublines ranged from 85 to 1500 ng/ml for Adriamycin and 65 to 10,000 nm for FdUrd. Some recurrences were significantly more resistant while others were more sensitive than the corresponding primary tumor lines. All recurrences had smaller 90% lethal dose values than the corresponding parent or primary tumor in response to Adriamycin; whereas, to 5-fluoro-2′-deoxyuridine, 90% lethal dose values revealed that most lines were quite resistant. Statistically significant differences in radiation survival were observed only for lines LR1a and LR5 (more sensitive). There was no apparent correlation between sensitivities to chemotherapy agents or X-irradiation and experimental metastatic potential in LR sublines. These dose-response data indicate that locally recurrent tumors are frequently, but not always, different from the original primary tumor in response to chemotherapy agents and ionizing X-irradiation. Although an exact mechanism is unknown, it is likely that “selective” pressures which eliminate large numbers of cells, in this case surgery, change tumor composition so that recurrent tumors may no longer be equivalent to the tumor mass that was originally excised. This suggests that treatment strategies should be planned accordingly.