ABSTRACT
On a journey through the Harz Mountains near Göttingen in winter 1777, Goethe observed the colors of the snow-covered landscape and described them as yellowish during daylight changing into a wonderful purple near sunset. The shadows cast on the snow appeared to him bluish to violet at first, turning into ‘green comparable in its clarity to green of the sea (Meergrün), but in its beauty to emerald green’ when the sun disappeared behind the horizon (Goethe, 1810, part I, § 75). The phenomenon of colored shadows must have fascinated Goethe for a long time, maybe since 1770, when he noted a book by Nikolaus von Béguelin on ‘Mémoire sur les ombres colorees’, published in Berlin 1767 (Ott, 1979). According to Béguelin (1767) and Monge (Kuehni, 1997) it seems that the first scientific description of the colored shadows goes back to an announcement by de Buffon in 1743, taking up an earlier observation by Abbé de Sauvages. In a preliminary study to his Farbenlehre Goethe described the colored shadows for the first time in 1792. Probably the first detailed and in many respects interesting paper was published by Gaspard Monge in 1789, drawn to our attention by John Mollon (1995). Finally, there is a contribution to the topic by Benjamin Thompson on ‘An account of some experiments upon coloured shadows’ in the Philosophical Transactions of the Royal Society in 1794. According to all three contributions of the late eighteenth century the most frequent and convenient way to observe colored shadows is described as follows: a white piece of paper or another white surface is illuminated by a candle and at the same time by dim daylight at dusk or dawn. An object will then cast two shadows, one by the candlelight, and one by daylight. The shadow cast by daylight is illuminated by the yellow candlelight and appears yellow. The shadow cast by the candlelight is illuminated by daylight, and although we expect it to appear grey or white this shadow is perceived as blue or blue–green.
