cap was 2,035 miles across. Inasmuch as the inclination of the Martian equator to the plane of the Martian orbit is, according to Schiaparelli, 24° 52', it must have then covered more than the whole south frigid zone of the planet.
Now, to take in the full meaning of the condition of the cap at this time and of the changes that ensued, we must begin by determining the Martian time of year. This is done by fixing the dates at which the Martian pole reached its maximum tilt toward or from the Sun, and the dates at which it was not tilted either to or from, but sideways to, the Sun; the former gives us the Martian solstices, and the latter the Martian equinoxes. It thus appears that on April 7, 1894, occurred the vernal equinox of the Martian southern hemisphere, on August 31, its summer solstice, and on February 7, 1895, its autumnal equinox. From these dates it is easy to transform the one calendar into the other. On the 3d of June, 1894, therefore, it was about May 1 on the southern hemisphere of Mars.
On May 1, then, Martian time, the cap was already in rapid process of melting; and the speed with which it proceeded to dwindle showed that hundreds of square miles of it were disappearing daily. As it melted, a dark band appeared surrounding it on all sides. Except, as I have since learned, at Arequipa, this band has