Accepting the accuracy of Schiaparelli’s discovery, which was corroborated in every particular in 1896 by Percival Lowell in a special series of observations on Mercury made with his 24-inch telescope at Flagstaff, Arizona, and which has also been corroborated by others, we see at once how important is its bearing on the habitability of the planet.  It adds another difficulty to that offered by the remarkable changes of distance from the sun, and consequent variations of heat, which we have already discussed.  In order to bring the situation home to our experience, let us, for a moment, imagine the earth fallen into Mercury’s dilemma.  There would then be no succession of day and night, such as we at present enjoy, and upon which not alone our comfort but perhaps our very existence depends, but, instead, one side of our globe—­it might be the Asiatic or the American half—­would be continually in the sunlight, and the other side would lie buried in endless night.  And this condition, so suggestive of the play of pure imagination, this plight of being a two-faced world, like the god Janus, one face light and the other face dark, must be the actual state of things on Mercury.

There is one interesting qualification.  In the case just imagined for the earth, supposing it to retain the present inclination of its axis while parting with its differential rotation, there would be an interchange of day and night once a year in the polar regions.  On Mercury, whose axis appears to be perpendicular, a similar phenomenon, affecting not the polar regions but the eastern and western sides of the planet, is produced by the extraordinary eccentricity of its orbit.  As the planet alternately approaches and recedes from the sun its orbital velocity, as we have already remarked, varies between the limits of twenty-three and thirty-five miles per second, being most rapid at the point nearest the sun.  But this variation in the speed of its revolution about the sun does not, in any manner, affect the rate of rotation on its axis.  The latter is perfectly uniform and just fast enough to complete one axial turn in the course of a single revolution about the sun.  The accompanying figure may assist the explanation.

[Illustration:  DIAGRAM SHOWING THAT, OWING TO THE ECCENTRICITY OF ITS ORBIT, AND ITS VARYING VELOCITY, MERCURY, ALTHOUGH MAKING BUT ONE TURN ON ITS AXIS IN THE COURSE OF A REVOLUTION ABOUT THE SUN, NEVERTHELESS EXPERIENCES ON PARTS OF ITS SURFACE THE ALTERNATION OF DAY AND NIGHT.]

Let us start with Mercury in perihelion at the point A.  The little cross on the planet stands exactly under the sun and in the center of the illuminated hemisphere.  The large arrows show the direction in which the planet travels in its revolution about the sun, and the small curved arrows the direction in which it rotates on its axis.  Now, in moving along its orbit from A to B the planet, partly because of its swifter motion when near the