1. Details have been given of a modified Stiven test for determining the efficiency of pasteurization in milk.
2. A ‘long’ form and a ‘short’ form of the test have been described.
3. The ‘long’ and ‘short’ tests have been compared with the Aschaffenburg & Mullen and Kay & Graham tests.
4. The work has led to the following conclusions:
(a) The ‘long’ Stiven test seems best to distinguish between exposure for 24 min. and exposure for 30 min. to 62·5° C, though compared with the other three tests this ‘long’ Stiven test as assessed in the present work is possibly too stringent.
(b) There was little difference between the results of the ‘short’ Stiven test, the Aschaffenburg & Mullen and the Kay & Graham tests when used for the routine examination of specimens.
(c) The ease of performance and the stability of the reagents make the Stiven test convenient to use, and it should be helpful to those who have difficulty in matching blues and yellows.
(d) It should be of value as a confirmatory procedure.