The break-up sequence of Gondwana has played a major role for the distribution of the biodiversity in the Southern Hemisphere, some leading to vicariance events. Vicariance biogeography uses most parsimonious areagrams in order to find evidence and explain such patterns. One notion is that areagrams convey biogeographic information to the extent that alternative palaeogeographic hypotheses are suggested. But different monophyletic groups often yield incongruent areagrams, not supported by geological data. Extinctions and dispersals are two important parameters which may distort the biogeographic information and lead to an incomplete biogeographic signal, and hence, flawed areagrams. Another parameter, often proposed by vicariance biogeographers, is that the continental reconstructions may be wrong. By reconciling a phylogeny with hypotheses of area relationships, it is possible to test hypotheses of dispersals and predict in which area(s) a particular group once were present. Well-preserved fossils are here a strong instrument, optimised to the reconciled tree, and used as evidence of past distributions. Nothofagus (Nothofagaceae) demonstrates that the alternative palaeogeographic hypotheses, suggested in the literature, fail to explain numerous past distributions in areas such as Antarctica, South America, and Tasmania. The only palaeogeographic hypothesis that explains all past distributions, is the current view of Gondwana break-up, supported by geological data. We therefore propose that Nothofagus may be used as a biological constraint on Gondwana break-up.