When fairly looked at, density would really seem to have a limit, except in so far as it may be combined with heat. We know that water is compressed 0·00005 part of its volume for every atmosphere of pressure to which it is subjected. But 0·00005 for round numbers, is in fractional numbers 1/20,000; therefore a pressure of 20,000 atmospheres would compress a cubic foot of water into 1/20,000 of a foot in height, or practically into nothing. We know, also, that as a column of water 33·92 feet high balances one atmosphere, one mile in height will be equal to 155·66 atmospheres, and 20,000 atmospheres will produce a pressure equal to a column of water 128 miles high; therefore, a cubic foot of water, subjected to such a pressure, would be compressed into virtually nothing. Again, supposing that we have a column of liquid rock, of 2-1/2 times the density of water, of the same height of 128 miles, we should have a pressure of 2-1/2 times that of the column of water; and as we have no reason to believe that granite in a liquid state has to obey a different law of compression to the one obeyed by liquid ice; then a column of granite 51 miles high would be sufficient to squeeze its own base, not only off the face of the earth but out of the bowels thereof. It will be seen, therefore, that at 100 miles deep from the surface, the density of the earth might well be equal to not only 5·66 times the density of water but to a great deal more; and that our estimate of 3 times the density of water, at 9 miles deep, was far within the mark.