Instead of acting merely as a guide-post, to _change_ the _direction only_ of the rays of heat, as before, it now becomes a receiving depot, absorbing nearly all the heat that comes to it. It must soon become filled, however. The thermometer hanging at the back has risen six degrees already, and is going up rapidly; it must soon begin to distribute its extra stores. But mark the different manner of distributing the heat. Instead of _reflecting_ the whole all in one direction, as when received on the other side, it now _radiates_ them equally in every direction. Some solid substances allow the rays, both of heat and light, to pass directly through them without either reflecting or absorbing them. Other substances allow the rays of light to pass through them, but absorb much of the radiant heat, like clear glass. Rock salt is one of the best non-absorbents of radiant heat, allowing nearly the whole of the rays of heat to pass through unobstructed. We will now return to our experiment at the other end of the tube. I find there is something wrong here--the mercury in the thermometer has risen several degrees. I knew this was rather a crude arrangement for illustrating this very beautiful and interesting part of our subject, but I hoped it would assist me a little in conveying to you the idea I desired to impress upon your minds. I find, however, that it is scarcely delicate enough to illustrate perfectly what I wanted to show. But this increased temperature is not owing to the effect of radiant heat on the air coming from the far end, for I find by the heat at the top of the pipe, between the heated ball and this ascending pipe, I, and by the current of heated air on the side next the ball, that there is a current of _circulating air_ that _has been heated_ by coming into immediate _contact_ with the hot ball. I designed this smaller tube, _k_, to carry off the air thus heated, but it appears to be too small. We ought to have had a piece of rock-salt to have closed the end of this tube, so that the radiant heat would have passed through without allowing any _circulation_ of _heated air_, but I was unable to find such a piece. But Professor Tyndall, in his lectures before the Royal Institute of Great Britain, gives the results of a large number of very accurate and beautiful experiments tried for the purpose of determining whether the forty-five miles of atmosphere surrounding the earth absorbed _any_ of the sun's rays, and if so, how much? These experiments prove, in the most conclusive manner, that dry pure air is almost a perfect non-absorbent of radiant heat. Thus, were the air entirely dry and pure, the whole forty-five miles through which the sun's rays have to pass, would absorb a very small fraction thereof, so that in the length of our tube it would be but an exceedingly small fraction of one degree, that is, for pure dry air. But is the air of this room pure and dry? Very far from it.