Raw milk is a well-established vehicle for the carriage of human pathogens, and many regulatory bodies have consequently mandated compulsory pasteurization as a food safety intervention. The residual activity of alkaline phosphatase (ALP) has historically been used to verify the adequacy of pasteurization of cow's milk. However, there is uncertainty on how the current ALP standards and methods of analysis can be applied to sheep's and goat's milk, which naturally contain different levels of the enzyme than that found in cow's milk. The official ALP methods applied in Canada (colorimetric assay; MFO-3) and in the United States (Fluorophos) were assessed for their ability to detect enzyme activity in raw and pasteurized milk derived from cows, sheep, and goats. The detection limit and the limit of quantitation were 0.8 and 2.02 microg/ml phenol, respectively, for the MFO-3 method and 43 and 85 mU/liter, respectively, for the Fluorophos method. The average ALP levels in raw goat's, cow's, and sheep's milk were 165, 1,562, and 3,512 microg/ml phenol, respectively. Raw milk detection limits, which correspond to raw milk phosphatase levels, were 0.051, 0.485, and 0.023% in cow's, goat's, and sheep's milk, respectively, for the MFO-3 method and 0.007, 0.070, and 0.004%, respectively, for the Fluorophos method. Although both methods can be used for ALP determination in cow's, goat's, and sheep's milk, the Fluorophos assay was superior to the colorimetric MFO-3 method based on sensitivity and time required to complete the analysis.