We recently described a new method for measuring the carcinogenicity of mineral oils, etc. (J. Hygiene, 35, 125). Further researches have provided a considerable amount of fresh data, which are briefly discussed in this paper. It may be recalled that the basic principle of the test depends upon the fall in the refractive index of an oil when injected into, and subsequently recovered from, the peritoneal cavity of the experimental animal. The standard procedure we have provisionally adopted is the injection of 0·5 c.c. of oil into a 20-g. male albino mouse, the oil being recovered after remaining 1 week in the animal. Up to the present we have performed about 800 tests with some 200 oils, half of which were ordinary commercial products. The general indications we have obtained are that the fall in the refractive index (K.I.F.) is proportional to the degree of unsaturation or dehydrogenation of an oil of given viscosity, the fall being a measure of the carcinogenicity plus dermaticity of the oil in question.