One theory seems to have a good deal of cogency. It depends on three known alterations in the relation of the earth to the sun. The first is a slow, regular change in the shape of the earth’s orbit through a cycle of 92,000 years. The second is a shift in the inclination of the earth’s axis through 40,000 years. The third is what a layman would call the wobble of this axis through 21,000 years. The first change increases or decreases the distance of the earth from the sun. The other two alter the angle of the sun’s rays and thus also increase or decrease the warmth given a particular area of the earth at certain seasons. No single unfavorable position would have had a great deal of effect in lowering summer temperature in the northern hemisphere, but two occurring at the same time—let alone three—would have appreciably diminished the sun’s heat. This astronomical theory of the cause of glaciation goes back a hundred years. As long ago as 1842 the French mathematician and astronomer J. Adhémar suggested that changes in the earth’s axis increased rainfall and provided the floods which he thought had moved the erratic blocks. Between 1864 and 1875 James Croll combined the wobble of the earth’s axis and the change in the earth’s orbit. A number of other men worked unsatisfactorily on the problem. The Serbian astronomer and physicist Milutin Milankovitch combined all three, and, between his first publication in 1913 and his latest in 1938, calculated the variations of solar radiation for the past 650,000 years.[8] In 1924 W. Köppen and A. Wegener applied Milankovitch’s early figures to the glaciation question, and Frederick E. Zeuner has lately used the revised figures of Milankovitch. Zeuner’s results, somewhat simplified, appear on page 55. They are fairly close to the geological estimates of B. Eberl and W. Soergel; his last two glaciations extend further back than those of Penck and Brückner.[9] Zeuner’s dates do not agree, of course, with those of an extremist like the geologist Kirtley F. Mather, who dates the first, or Günz, glaciation as ranging from 2,000,000 to 1,500,000 years ago.[10] Zeuner’s findings work out well enough for the American glaciations, except that there is no New World equivalent for his first Würm maximum of 115,000 years ago. Many authorities refuse to accept any such condition in the Old World. Because of this glaciation Zeuner moves back the appearance of _Homo sapiens_ a good 50,000 or even 75,000 years. [Illustration: THE FOUR GREAT GLACIATIONS _Six varying estimates of their duration made by five authorities. Fisk’s are of the New World glaciers, which are generally equated with those of Europe._]