The tumor-producing effect of automobile exhaust condensate and fractions thereof. Part III: mathematical-statistical evaluation of the test results.

This paper deals with the mathematical-statistical evaluation of experiments devoted to the following problems: 1. To what extent is the carcinogenic potency of automobile exhaust gas condensate dependent on the dose? 2. What fractions (groups of substances) of the automobile exhaust gas condensate are mainly responsible for the carcinogenic potency? 3. To what extent can the potency be explained by fractions (or single substances)? In order to answer the above questions, the working group planned animal experiments (cf. Misfeld and Timm, 1973) and obtained, prepared, and fractionalized automobile exhaust gas condensate (Grimmer, 1978). The exhaust gas condensate and its fractions were dropped onto the skin of female CFLP mice (Brune et al., 1978). In view of the problems posed above, the aim of the mathematical-statistical evaluation was to establish the dose-response relations, to assess the combined effect, and to estimate the relative potency and the assumed potency. In particular, the hypothesis (basis for the separation process) that the potency of the entire exhaust gas condensate is mainly attributable to the fraction which contains the polycyclic aromatic hydrocarbons (nitromethane phase) was examined.