Mitosis counting remains one of the most valuable prognostic indicators in tumor pathology; however, as currently carried out it is time consuming and not reproducible. In this study, 6 different pathologists, using different microscopes, arrived at widely different mitotic counts on the same slide, ranging from 4 to 16. These differences were mainly due to the different field areas of the various microscopes used and the method used for counting and recording. In evaluating the most active 10 HPF, the count ranged from 10 to 19. Instead, when an average of 40 fields was recorded, the range was 4-11. Using the mitosis/volume index, which expresses the number of mitotic figures per mm2 of viable tumor, the counts ranged from 8 to 10, a marked improvement. However, this method is complicated and not "user-friendly.'' We suggest a variation of the technique by which a 2 mm2 rectangle is drawn on a cover slip and mounted under the microscope, centered on the most mitotically active area of the tumor. The mitoses in that area are counted (=n) and the percent of viable tumor (=x%) is estimated under low magnification. The number of mitoses per mm2 of viable tumor (cs-MAI) is then calculated according to the formula Cs-MAI=100n/2x. Using this modified method, the range of mitoses counted by the different observers was very narrow (9 to 11), and the time required for the counting was only 5-10 minutes.