-1859] WRITINGS OF JOSEPH HENRY. 157 



that the total height of an atmosphere surrounding a planet 

 does not depend upon the quantity of gas of which it is 

 constituted. To prove this, let us imagine a vertical column, 

 say an inch square at the base, filled with air of a given 

 density extending to the top of the atmosphere. Let us 

 suppose this column to be divided into portions an inch 

 high throughout its whole length b}^ movable planes, and 

 into each one of these portions double the quantity of air to 

 be introduced. The lowest portion, namel}', the first inch, 

 will not be enlarged by this condition; for though twice as 

 many repellant atoms are introduced into the same space, 

 tending to repel upward the first dividing plane, yet this 

 plane will be pressed downward by twice the weight, because 

 twice the number of atoms have been introduced into all 

 the strata above. 



The same reasoning may be applied to all the successive 

 strata until we come to the very highest. On this no addi- 

 tional weight is placed, and it would therefore expand until 

 the diminution of its elasticity just equals its own weight, 

 and at this point the equilibrium will take place. If how- 

 ever this point should be just at the place of equilibrium 

 where the weight of the atom would be overcome by the cen- 

 trifugal force, the upper film would be removed, another 

 would expand into its place, and another, and another, until 

 the whole atmosphere would be withdrawn. This, as we 

 have said, is a logical consequence of the extension of the 

 law of Mariotte, and has been applied by Dalton and others 

 to determine the heights of mixed atmospheres, or of atmos- 

 pheres of different densities. But the height of the atmos- 

 phere is probably far below the point where the weight of 

 the atom is equal to the force of gravity, since this may be 

 found by calculation to be at about 5'6. times the earth's 

 radius from the surface at the equator, or about 22,400 miles. 

 If we suppose the column to be formed of a lighter gas, as 

 for example hydrogen, the atoms of which have the same 

 repulsive energy as those of air, then the column will be in- 

 versely proportioned to the density at the surface, and from 

 this we can readily calculate the relative heights of atmos- 



