Variations in precision of the Breed method for cell counts in milk were investigated by utilizing different working factors (WF) for the same smear and by using the same WF over a major portion of the probable cell count range. A significant inverse relationship was found between precision and the WF. With a constant WF on the other hand, the precision of the count increased very significantly as the actual cell count increased. Formulas showing the relationship between the expected high and low for any given cell count were computed via the least squares method for a WF of 20,000. Evidence was presented that a WF of 5,000 or below would be necessary when a good estimate of cell content is important.
1This work was part of a thesis submitted by R. Schneider to the Graduate Division, University of California, in partial fulfillment of the requirements for the M.S. degree.