They are of various depths, and they may be sometimes observed to communicate with each other. Their artificial appearance was sufficient to convince of the truth of his system regarding them, this strenuous supporter of a worship which must on his hypothesis have required a priesthood sufficient to exclude all other population, if every rounded cavity which the granite exhibits was a pool of lustration.
Their true origin is easily traced by inspecting the rocks themselves. On examining the excavations, they will always be found to contain distinct grains of quartz and fragments of the other constituent parts of the granite. A small force is sufficient to detach from the sides of these cavities additional fragments, showing that a process of decomposition is still going on under favourable circumstances. These circumstances are the presence of water, or the alternate action of air and water. If a drop of water can make an effectual lodgment on a surface of this granite, a small cavity is sooner or later produced. This insensibly enlarges as it becomes capable of holding more water, and the sides as they continue to waste, necessarily retain an even and rounded concavity, on account of the uniform texture of the granite. In time, the accumulated gravel is blown away by the winds, although in the deeper hollows it may often be found forming considerable accumulations.
The same solubility of granite in water, (to speak generally) is the cause of that wasting of the surface which these rocks undergo, and to which I have before attributed the enlargement of the vacuities at the surfaces of contact, and the separation of the prisms into detached masses.
We need not hesitate in admitting the solution of granite in water to an extent capable of producing this effect of disintegration, since we know that silex is soluble in that fluid by natural means, however we