THE demonstration relating to the density and depth of the solar atmosphere, published in NATURE October 5, 1871, page 449, has been entirely misconceived by Mr. Ball. The volume of the terrestrial atmosphere is an element which obviously has nothing to do with the question. Atmospheric air, if raised to a temperature of 3,272,000° Fah., will expand 6,643 times; hence a vertical column forty-two miles high will reach a, height of 279,006 miles, if brought to the stated temperature. The basis of computation adopted by Captain Ericsson being an area of one square inch, he shows that a medium similar to the terrestrial atmosphere containing an equal quantity of matter for corresponding area, transferred to the solar surface, will, owing to the superior attraction of the sun's mass, exert a pressure of 14.7 X 27.9 = 410 pounds. And that, if the said medium be heated to a mean temperature of 3,272,000° Fah., it will expand to a height of 279.006/27.9 = 10,000 miles above the solar surface. But. if a gas composed chiefly of hydrogen 1.4 times heavier than hydrogen the specific gravity of which is 1/14 of that of air, be substituted, the height will be (14 X 10,000)/1.4 = 100,000 miles. Admitting that the ascertained coefficient of expansion, 0.00203 for 1° Fah., holds good at the high temperature before referred to, the stated altitudes of the solar atmosphere cannot be disputed. Mr. Ball's announcement concerning the properties of spheres, it is scarcely necessary to observe, has no bearing on the foregoing calculations. With reference to the effect of intense heat, it will be well to bear in mind that the before-mentioned rate of expansion holds good for atmospheric air—within an insignificant fraction—under extreme rarefaction as well as under high temperatures. We have no valid reason, therefore, to suppose that any deviation from the ascertained law of expansion takes place in the solar atmosphere, sufficient to alter materially the before-mentioned computations of its depth.