The bluish tint of clays is due to finely divided iron pyrites (iron disulphide), which may occasionally appear as distinct crystals or nodules of one or other of its forms, pyrite or marcasite. On oxidation, limonite arises, which colours the mass brown, as is seen in the upper part of many clay-pits. The occurrence of iron pyrites often dates back to the time at which the clay accumulated. N. Andrussow[42] points out that in the Black Sea there is an enormous supply of decaying organic matter provided by the floating organisms of the upper layers. This rains continually down towards the floor. The portion that reaches depths of over 100 fathoms escapes from the voracity of free-swimming organisms and arrives at the region where bacteria alone abound. These bacteria act on dissolved sulphates, and also largely, according to Andrussow, on the albumen of the decaying matter. In both cases, sulphuretted hydrogen is produced. Andrussow treats the reduction of the marine sulphates as a minor process, due to the need that the bacteria have for oxygen in the deep waters, which are insufficiently supplied. The sulphuretted hydrogen attacks the salts of iron, and iron disulphide results. Here we have an excellent illustration of how, in deep basins, with imperfect vertical circulation, black pyritous muds may arise, devoid of ordinary fossils. The depths of the Black Sea are practically poisoned by the abundance of sulphuretted hydrogen. But numerous cases of shales are known to us where iron pyrites replaces the shells of ammonites or forms complete casts of bivalves, and has accumulated also in concretions and crystalline groups. Such pyrites is probably of secondary origin, or arose from the reducing action of decaying organic matter on ferrous sulphate in solution in the sea. The oxidation of iron pyrites in shales gives rise to aluminium sulphates, such as alums. Sometimes sufficient heat is evolved during this oxidation to set on fire carbonaceous matter present in the rock. Pink-purple and green are common colours among shales, and imply that the iron is in two different states of oxidation. When the colour varies thus in successive bands, we may believe that a climatic change promoted the formation of ferric salts on the land surface when the pink layers were being formed, while ferrous (less oxidised) salts predominated when the green particles were washed into the basin. B. Smith[43] suggests that the organic matter and humic acids which are swept down in times of flood may temporarily prevent oxidation from occurring in shallow lakes and pools. Dry seasons would thus lead to the deposition of pink clays, while wet seasons would furnish green ones. The green colour in shales is mostly due to chlorite or to glauconite.