To illustrate the amazing productiveness of the South African mines, it may be mentioned that, according to Gardner F. Williams, the Kimberley group of mines in sixteen years yielded 36 million carats of diamonds, and the annual output of the Jagersfontein mine averages about a quarter of a million carats, whereas the total output of the Brazil mines, for the whole of the long period during which they have been worked, barely exceeds 13 million carats. The average yield of the South African mines, however, perceptibly diminishes as the depth of the mines increases. The most interesting point connected with the South African diamond mines, viewed from the scientific standpoint, is the light that they have thrown on the question of the origin of the diamond, which previously was an incomprehensible and apparently insoluble problem. In the older mines, just as at the river diggings by the Vaal, the stones are found in a gravelly deposit that has resulted from the disintegration of the rocks through which the adjacent river has passed, and it is clear that the diamond cannot have been formed _in situ_ here; it had been suspected, and now there is no doubt, that the itacolumite rock of Brazil has consolidated round the diamonds which are scattered through it, and that it cannot be the parent rock. The occurrence at Kimberley is very different. These mines are funnels which go downwards to unknown depths; they are more or less oval in section, becoming narrower with increasing depth, and are evidently the result of some eruptive agency. The Kimberley mine has been worked to a depth of nearly 4000 feet (1200 m.), and no signs of a termination have as yet appeared. The blue ground which fills these ‘pipes,’ as they are termed, must have been forced up from below, since it is sharply differentiated from the surrounding country rocks. This blue ground is a brecciated peridotite of peculiar constitution, to which the well-known petrologist, Carvil Lewis, who made a careful study of it, gave the name kimberlite. The blue colour testifies to its richness in iron, and it is to the oxidation of the iron constituent, that the change of colour to yellow in the upper levels is due. Owing to the shafts that have been sunk for working the mines, the nature of the surrounding rocks is known to some depth. Immediately below the surface is a decomposed ferriferous basalt, about 20 to 90 feet (6-27 m.) thick, next a black slaty shale, 200 to 250 feet (60-75 m.) thick, then 10 feet (3 m.) of conglomerate, next 400 feet (120 m.) of olivine diabase, then quartzite, about 400 feet (120 m.) thick, and lastly a quartz porphyry, which has not yet been penetrated. The strata run nearly horizontal, and there are no signs of upward bending at the pipes. The whole of the country, including the mines, was covered with a red sandy soil, and there was nothing to indicate the wealth that lay underneath. The action of water had in process of time