A selective history of the benzene problem is presented, starting with August Kekulé's proposal of a hexagonal structure in 1865 and his hypothesis of 1872 that the carbon–carbon bonds oscillate between single and double. Only those theories are included that were accepted or at least discussed by a significant number of chemists. Special attention is given to predictions, their empirical tests, and the effect of the outcomes of those tests on the reception of the theories. At the end of the (...) period covered by this article, chemists generally accepted the valence bond theory proposed by Linus Pauling; some of them considered this a more sophisticated version of Kekulé's oscillation hypothesis. The sequel describes the replacement of the valence bond theory by the molecular orbital theory in the period ending around 1980. (shrink)