Factors which might contribute to variance of cell estimates in milk, using the Direct Microscopic Somatic Cell Count ( DMSCC) were investigated, Comparisons of cell estimates obtained from counting cells in nine areas of 20 circular milk films produced no evidence of uneven distribution of cells on these films. Estimates of cells by DMSCC were not affected by the temperature of the milk, or the method of spreading, when milk films were prepared. Standard deviations and means of counts indicated no significant difference in cell estimates obtained by three operators. Cell estimates from counts made in horizontal and vertical diameter strips of milk films showed no significant differences among means or variances relating to strip direction. Cells were counted in diameter strips (as defined by a wide and narrow ocular reticle) at two magnifications (450 × and 1,000 ×). Cell estimates based on counts from the wide strips had lower standard deviations than those from the narrow strips.
Confidence limits were established for estimating 1 × 106 cells per ml of milk at p = 0.05.
A procedure for determining precision within laboratories and permitting a comparison between laboratories is presented.
1Present address: New York State Veterinary College, Cornell University, Ithaca, New York. 14850.