Let us consider what this theory really asserts. If the earth were a perfect sphere, its orbit round the sun a perfect circle, and the equator coincided with the ecliptic, there would be no seasons. The four quarters of the year would each receive the same quantity of solar heat and light, and the days and nights would be always equal. But the inclination of the equator to the ecliptic, that is, of the earth's plane of daily rotation to that of its annual revolution, necessitates seasons. Each pole must be alternately turned to and away from the sun every year. Each hemisphere, therefore, must have alternately its spring, summer, autumn, and winter. But if the earth's orbit were exactly circular, these seasons would be of equal duration, and the distance from the sun no greater in one than in another. But the earth's orbit is not circular, but elliptic, and the eccentricity, or deviation of the oval from the circular form, varies considerably over very long periods, though always coming back to the amount from which it started. These variations are due to perturbations from the other bodies of the solar system acting according to the law of Newton's gravitation, and therefore calculable. Again, the earth is not a perfect sphere, but a spheroid, and there is a factor called precession, due to the attraction on the protuberant mass at and towards the equator. The effect of this is, that instead of the earth's axis pointing uniformly towards the same celestial pole, it describes a small circle round it. This circle is completed in about 21,000 years, so that if the earth is nearest to the sun when the North Pole is turned away from it, and it is winter in the northern hemisphere, as is now the case, in 10,500 years the conditions will be reversed, and the southern hemisphere will be in perihelion, or nearest the sun, when its pole points away from it. And as the perihelion portion of the earth's orbit is, owing to its eccentricity, shorter and more quickly traversed than the aphelion portion, this means practically that winters will be shorter than summers in the hemisphere which precession favours, and longer in that to which it is adverse. As precession now favours the northern hemisphere, which is warmer than the southern in corresponding latitudes, it might be thought at first sight that this was the cause of the glacial period. But it is evident that this is not the case, for the precessional revolutions come round far too rapidly, and it is impossible to suppose that there have been glacial and genial periods alternating every 10,500 years, with all the inevitable changes of seas and lands, and of fauna and flora, accompanying each alternation throughout the whole of geological time. In fact, it is abundantly evident, on historical evidence alone, that there has been no approach to any such changes during the last 10,500 years, which carries us back to a period when our northern summers were short and our winters long.