If indeed a small spectrum be thrown on the screen and the different colours are examined with the thermopile, it will be found that as long as we allow it to remain at the blue end of the spectrum, there will be no effect on the galvanometer, but if instead of holding it at the blue end we bring it towards the red, the galvanometer needle is deflected from its normal position, to that it had when the red rays fell on it, showing that it is beyond all doubt the red rays and not the blue to which it is sensitive. Where then in the spectrum are the rays which affect the photographic plate? We can at once settle this point. If one be placed in the spectrum for a short time, and then developed, it will be found to be affected only in the part on which the blue rays have fallen. Indeed to demonstrate this no lamp is necessary. If for half-an-hour or so two pieces of sensitive paper are placed in the daylight, one covered with red glass, and the other with violet, so that the sunlight is made to travel, in the one case, through red glass, and in the other through violet, it will be found that the violet light will act, and produce a darkening of the paper, while the red glass will preserve the paper below it from all action. This is a proof that the blue end of the spectrum has another kind of energy, a chemical energy, by means of which certain chemicals are decomposed, this is the basis of photography. These different qualities of light have been utilized by the astronomer. He attaches a thermopile to his telescope and establishes a celestial thermometry. The radiations repay a still more minute examination, and aided by the spectroscope, he is able to study with the utmost certitude the chemical condition of the heavenly host, while the polariscope enables him to acquire information in still another direction. Nor does he end here. He replaces his eye by a sensitive plate, which not only enables him to inquire into the richness of the various bodies in these short waves, but actually to obtain images of them of most marvellous beauty and exactness. These various lines of work we have to consider in the remaining chapters. CHAPTER XXVI. DETERMINATION OF THE LIGHT AND HEAT OF THE STARS.