prise almost exactly eighty-four terrestrial days. According as the arc of solar oscillation is all above the horizon of the observer or all below it, or partly above and partly below it, there will be different appearances and a different distribution of light and heat. In the regions, covering three eighths of the planet, where the arc is all below the horizon, the sun will never be seen, and the darkness will be perpetual. Thick and eternal night will reign there, except perhaps from the accidental appearance of some light produced by refraction and atmospheric glows, or phenomena like the aurora borealis; together with the light emitted by the stars and planets.
Another part of Mercury, including also three eighths of its surface, will have the arc of oscillation all above its horizon, and will be continually exposed to the rays of the sun, without any other change than the variations in the obliquity of the rays through the different phases assumed during the period of eighty-eight days. Night is absolutely impossible. In other regions, covering a quarter of the planet, in which the arc of oscillation is partly above and partly below the horizon, there will be alternations of light and darkness. In these privileged regions the period of eighty-eight days will be divided into two intervals, one characterized by a continuous light, the other by darkness; the two intervals will be equal in some places, of different length in others, according to the position of the place on the surface of the planet, and the length of the part of the solar arc which appears above the horizon.
The possibility of organic life in a planet constituted after this manner depends on the existence of an atmosphere capable of distributing heat into different regions, in such a way as to diminish the extremes of heat and cold. Schroeter, a hundred years ago, suspected the existence of an atmosphere round Mercury; my observations afford more definite indications of it, and affirm its existence with a much greater probability. The spots of the planet are most clearly visible when they are in the central parts of the disk, and grow dimmer and ultimately disappear as they approach the border. I have been able to assure myself that this phenomenon is not merely due to the greater obliquity of the perspective, but is because some obstacle is really presented to the view of spots situated in such positions. That obstacle can hardly be anything else than the greater extent of atmosphere that the light-rays have to traverse in coming from the edges than from the center of the disk. We have, therefore, reasons for believing that the atmosphere of Mercury is less transparent than that of Mars, and more nearly like that of the earth. The circular contour of the planet, moreover, in which the spots become less visible, always appears more luminous than the rest, but often irregularly