We may get the beginning of what may be an explanation of all the facts from another part of Captain Ericsson's monograph, where he says: "Engineers of great experience in the application of heat for the production of motive power and other purposes deny that the temperature of a body can be increased by the application of heat of a lower degree than that of the body whose temperature we desire to augment." The soundness of their reasoning is apparently incontrovertible, yet the temperature of the mercury in the instrument just described raised to 600°F. by means of the parabolic reflector, increases at once when solar heat is admitted through the circular apertures, although the sun's radiant intensity at the time may not reach one-tenth of the stated temperature. It should be mentioned that the trial of this new pyrheliometer has not been concluded, owing to very unfavourable atmospheric conditions since its completion. For our present purpose the great fact established by the illustrated instrument is sufficient, namely, that the previous temperature of a body exposed to the sun's radiant heat is immaterial. The augmentation of temperature resulting from exposure to the sun, the pyrheliometer shows, depends upon the intensity of the sun's rays.
A little study shows us that the steam engineers are perfectly right in their doctrine. The heat of steam can only be called a variety of the temperature of water. At 300 lb. pressure per square inch the heat of steam is 417·5°F., while at 20 lb. pressure it is only 228·0°F., and therefore the steam engineer has good reason to say that steam at the lower pressure--or derived from heat that can only produce that pressure--can add no heat to the higher; on the contrary, the only possible means of applying the heat of the lower to that of the other would be by mixing them, and we know what the result of that would be. This brings before us the fact that the steam engineer's heat is very limited, and can only be communicated in certain ways, while the sun's heat is comparatively unlimited, and can only be communicated to anything through the medium of the ether. But it probably teaches more than that. Were the engineer's heat unlimited in quantity at low pressure it can easily be believed that it could be transmitted to another body at any temperature by radiation, the same as it is radiated from the sun to a hot box; but it is not, and we thus seem to find that radiation is a mode, possessed by the ether alone, of conveying heat from one body to another. It has nothing whatever to do with mixing, conduction, convection, or anything, except in so far as the ether is mixed in a more or less limited quantity with all matter. In support of this idea we can refer to Professor Tait's treatise on heat, where we find it stated that "heat does pass (though on an infinitesimal scale) from colder to hotter bodies"; and we can easily understand that the infinitesimal quantity so passed is due to the comparatively infinitesimal quantity of ether there is in either of the two bodies to perform the work of transference. Professor Tait has not told us how heat is carried from a cold to a hot body, but there can be no doubt about its being a function of the ether which can only be found out by a careful and analytical study of that agent. Such a study we propose to undertake presently without much expectation of being successful, but still with the hope of helping in some measure to find out how the ether operates. Meanwhile we shall return to what we had begun to say about the sun being a hollow sphere, and to our proposal to treat of the nebula contracted from 58 million miles to its present diameter, as if it were a model representing a _résumé_ of all the effects produced on the nebula by that amount of condensation.