of only one member of the Carboniferous system, itself only a member of the great Palaeozoic group, and we have made no allowance for the abrasion from previous rocks and deposition of the immense mass of sandy and muddy sediment in which the coals and forests are imbedded, and which is vastly greater than the deltas of the largest modern rivers. Considerations of a physical rather than of a geological nature also give us long periods for the probable existence of the earth, though they serve to correct somewhat the extravagant estimates of some theorists. Croll has based an interesting calculation on the amount of erosion of the land by rivers. That of the Mississippi amounts to one foot in 6000 years. That of the Ganges gives one foot in 2358 years, the average being, say, one foot in 4179 years. Some smaller rivers give a much shorter time; but the average of two great rivers, one draining a very large area of the western and another of the eastern hemisphere, and in very different climates and geographical conditions, will probably be the most reliable datum. Croll, however, prefers the Mississippi rate.[145] If we estimate the proportion of land to water as 576 to 1390, this will give for the entire area of the ocean a rate of deposition of one foot in 14,400 years. Now the entire thickness of all the stratified rocks is estimated at 72,000 feet; and at this rate the enormous time of 1,036,800,000 years would be necessary. But we have no right to assume that deposition has been going on uniformly over the entire sea-bottom. On the contrary, the greater part of it takes place within a belt of about one hundred miles from the coasts, and the deposit of calcareous and other matters over the remainder will scarcely make up for the portions of this belt on which no deposit is taking place. This will give an area of deposit of about 11,650,000 square miles, consequently only one twelfth of the above time, or about 86,400,000 years, would be required. This can be but a very rough calculation; but it has the merit of squaring very nearly with the calculations derived from physical considerations, more especially by Sir William Thomson, which limit the possible existence of the earth's solid crust to one hundred millions of years. Similar conclusions have also been deduced from what is known of the physical constitution of the sun. Croll's own ingenious theory of glacial periods produced by the varying eccentricity of the earth's orbit, along with the precession of the equinoxes, would give, according to him, about 80,000 years ago for the date of the Glacial period, and for the beginning of the Tertiary period about 3,000,000 years ago.