Proliferative activity has been proposed as a prognostic and predictive marker for breast cancer; Ki-67 is one of the most frequently used markers to assess proliferative activity. In the current study, Ki-67 immunoreactivity was comparatively assessed, even in terms prognostic relevance, with 3H-thymidine labeling index as a reference standard for proliferation in 126 patients with stage I and II breast cancer. There was a significant but weak correlation between Ki-67 values and the 3H-thymidine labeling index (r = 0.19, P = 0.03). Analysis of variance showed that the mean 3H-thymidine labeling index values were not statistically different in terms of pathologic size (T1, T2. T3, T4), number of pathologically positive axillary nodes (neg, pos 1-3, pos > 3), and grading classes (1, 2, 3), but significantly and inversely correlated with estrogen receptor status (P = 0.033) and progesterone receptor status (P = 0.08). The Ki-67 values significantly correlated with N status (P = 0.041), estrogen receptor status (P < 0.001), progesterone receptor status (P < 0.001), and grading (P < 0.001). The median follow-up was 37 months. In terms of prognosis, Ki-67 was associated significantly with overall survival (P = 0.01) and marginally with disease-free survival (P = 0.095). A significant difference in prognosis was found for both disease-free survival (P = 0.024) and overall survival (P = 0.040) when a 3H-thymidine labeling index cut-off of 6.5% was used (P = 0.024). The results suggest that, although both are indicators of proliferative activity, 3H-thymidine labeling index and Ki-67 seem to identify breast cancers with different phenotypes.