Abstract Throughout his life, Lyell travelled extensively, always as a keen observer. He viewed the Earth's geological history as continuous with and subject to the same processes of change as at present. Leopold von Buch's theory of craters of elevation contradicted Lyell's view of Earth history. Thus Lyell travelled to Madeira and the Canary Islands in 1853 to see von Buch's evidence. Lyell found the islands formed by a long series of volcanic eruptions, not by the single explosive upheaval that von Buch had described. Nevertheless, Lyell still accepted Léonce Élie de Beaumont's claim that lava flows could not form compact rock on steep slopes. In 1855, Lyell learned from Eilhard Mitscherlich that on Stromboli contemporary steeply inclined lava flows were forming solid rock. In 1857, Lyell went to Sicily where unmistakable evidence contradicted Élie de Beaumont. In the walls of the Valle del Bove, steeply inclined layers of lava were intersected by dykes that pointed towards a former centre of eruption at Trifoglietto, later buried by volcanic rocks emitted from the present centre of eruption at the summit of Etna, proving that the Valle del Bove could not have originated as a crater of elevation.