level; but in most cases they have originated from the atmospheric agencies that I have mentioned, attacking most destructively the softer portions of the rocks, which are usually of a calcareous nature.
The caverns of the other class also generally occur in limestone districts, and seem in like manner to be mainly due to atmospheric causes, though operating in a different manner. They usually appear to have originated with some small crack or fissure in the rock, along which, water falling on the surface was able to find its way to some vent at a lower level; and this, by its continual passage, was able to enlarge the channel along which it flowed. The mechanically erosive force of pure water in passing over or even falling upon a rock of moderate hardness is indeed but small, though its powers of friction were long since recognized by that most enlightened of ancient geologists, the poet Ovid,[1] who classes its effects with the wearing away of a ring upon the finger. Nor was Solomon's likening of the contentions of a wife to a continual dropping, without its geological significance. But in the case of water derived from rain falling on the surface, and passing through a fissure in a limestone rock, its first effects are chemical rather than mechanical.[2]
By contact with decaying vegetable matter the water becomes charged with a certain amount of carbonic acid, and is rendered capable of dissolving a portion of the calcareous rock through which it passes, and thus carries it off in solution, while in so doing it acquires the character known as "hard." Taking the case of water delivered by springs in the chalk, which has but a moderate degree of hardness, it is proved by analysis to contain about seventeen grains of carbonate of lime to the gallon. Now, out of a rainfall of say twenty-six inches annually, it has been found by experiment, that in a chalk district about nine inches would, in average seasons, make their way down to the springs;. and it may be readily calculated that at the rate of seventeen grains to the gallon, the amount of dry chalk or carbonate of lime dissolved by this quantity of water, and delivered by the springs, and thus carried away, is, in each square mile of such a district, upwards of one hundred and forty tons in each year, or about a
