Modification of Babcock Method to Eliminate Fat Testing Bias Between Babcock and Ether Extraction Methods (Modification of AOAC Official Methods 989.04 and 995.18): Collaborative Study

Abstract

Abstract The Babcock test for determining fat in milk (AOAC Official Method 989.04) and in cream (AOAC Official Method 995.18) gives consistently higher test results than does the modified Mojonnier ether extraction reference method (AOAC Official Methods 989.05 and 995.19). To decrease the density of material in Babcock columns and thus lower test results, the Babcock method was modified by lowering the temperatures used at various points in the method, from about 57.5° to 48°C. Using the AOAC collaborative study format, 9 laboratories tested 9 pairs of blind duplicate raw milk samples (fat range 2.5–5.7%) and 9 pairs of blind duplicate heat-treated cream samples (fat range 30–45%) using the temperature modified (TM) Babcock method. The ether extraction test was conducted as the reference method. The statistical performance (invalid and outlier data removed) of the TM Babcock method was, for milk: percent fat value of 4.110, repeatability standard deviation (sr) value of 0.037, reproducibility standard deviation (sR) value of 0.047, repeatability relative standard deviation (RSDr) value of 0.901% and reproducibility relative standard deviation (RSDR) value of 1.147%; and for cream: percent fat value of 37.555, sr value of 0.258, sR value of 0.353, RSDr value of 0.687% and RSDR value of 0.940%. The TM Babcock method performance was acceptable but not as good as that achieved in previous studies of the unmodified method. For the TM Babcock and ether methods, respectively, average percentages fat were 4.110 and 4.114 for milk, and 37.555 and 37.485 for cream. Temperature modification statistically eliminated the testing bias between methods. The tem-perature modifications of the Babcock methods for determination of fat in milk and cream (989.04 and 995.18) have been adopted revised first action by AOAC INTERNATIONAL.