We know from all our work that the sun must be a gasiform body, which means that all the cosmic matter contained in it must be in the form of vapour, even although its consistence should outrival a London fog--notwithstanding that some physicists have supposed that it may be solid at the centre through extreme pressure--and it is not altogether correct to compare its construction to that of a solid body such as the earth; but as we have no other we shall begin to make a comparison with it, which, it will be found, can lead us into no appreciable error. Considering then the sun to be 867,000 miles in diameter, with mean density of 1·413 that of water, the hollow part being still completely empty, and applying to it the same proportion we have deduced for the earth, we find that the region of greatest density would be at 0·7937 of the radius of the sphere--a proportion really derived from the line of division into two equal parts of the volume of a sphere--from the centre, or 89,431 miles from the surface; and the inner surface of the shell at 0·548--a proportion derived from our calculations for the earth--of the radius of 433,500 miles, or 237,558 miles from the centre; which in turn makes the shell to be 195,942 miles thick, and the hollow centre to be 475,116 miles in diameter. On the other hand, still following the proportions derived from the earth, we find that the density at the surface might be one-third of the mean density or 0·471; that it might be one-fifth greater than the mean, or 1·7 at the region of greatest density and one-half, or 0·71 at the inner surface of the shell--all of these three densities being in terms of water.