Résumé : The carcinogenic activity of 4′,9,10‐trimethyl‐1,2‐benzanthracene is discussed in the light of the theories on constitution and carcinogenic activity put forward by Coulson, Greenwood, Badger, Pullman and Daudel. It is pointed out that the carcinogenic activity of this compound, substituted in the «benz‐ring» (4′‐position), invalidates the hypothesises of Badger and of Greenwood. 2′,9,10‐trimethyl‐1,2‐benzanthracene has been synthesised in a search for new carcinogenic derivatives of 1,2‐benzanthracene substituted in the «benz‐ring». This compound (m. p. 107‐107, 5°) was obtained by reacting 2′‐methyl‐1,2‐benzanthraquinone with CH3MgI, followed by treatment with HI+HBr and finally SnCl2+HCl (Sandin‐Fieser reaction). The U.V. spectra of 2′,9,10‐trimethyl‐1,2‐benzanthracene and bis (2′,9‐dimethyl‐1,2‐benzanthracenyl‐10‐)‐1,2‐ethane were determined. Copyright © 1955 Wiley‐VCH Verlag GmbH & Co. KGaA, Weinheim