There is thus in these great continental changes a law of recurrence and a law of progress; but as to the efficient causes of the phenomena we have as yet little information. It seems that original fractures and shrinkages of the crust were concerned in forming the continental areas at first. Once formed, unequal burdening of the earth's still plastic mass by deposits of sediment in the waters, and unequal expansion by the heating and crystallization of immense thicknesses of the sediment, may have done the rest; but the results are surprisingly regular to be produced by such causes. We shall also find that similar cycles can be observed in the geological ages which succeeded the Palæozoic. Geologists have hitherto for the most part been content to assign these movements to causes purely terrestrial; but it is difficult to avoid the suspicion that the succession of geological cycles must have depended on some recurring astronomical force tending to cause the weaker parts of the earth's crust alternately to rise and subside at regular intervals of time. Herschel, Adhémar, and more recently Croll, have directed attention to astronomical cycles supposed to have important influences on the temperature of the earth. Whether these or other changes may have acted on the equilibrium of its crust is a question well worthy of attention, as its solution might give us an astronomical measure of geological time. This question, however, the geologist must refer to the astronomer. [Illustration: CURVES SHOWING THE SUCCESSIVE ELEVATIONS AND DEPRESSIONS OF THE AMERICAN CONTINENT, IN SEVERAL CYCLES OF THE PALÆOZOIC TIME.] There are two notes of caution which must here be given to the reader. First, it is not intended to apply the doctrine of continental oscillations to the great oceanic areas. Whether they became shallower or deeper, their conditions would be different from those which occurred in the great shallow plateaus, and these conditions are little known to us. Further, throughout the Palæozoic period, the oscillations do not seem to have been sufficient to reverse the positions of the oceans and continents. Secondly, it is not meant to affirm that the great Permian plications were so widespread in their effects as to produce a universal destruction of life. On the contrary, after they had occurred, remnants of the Carboniferous fauna still flourished even on the surfaces of the continents, and possibly the inhabitants of the deep ocean were little affected by these great movements. True it is that the life of the Palæozoic terminates with the Permian, but not by a great and cataclysmic overthrow. We know something at least of the general laws of continental oscillations during the Palæozoic. Do we know anything of law in the case of life? The question raises so many and diverse considerations that it seems vain to treat it in the end of a chapter; still we must try to outline it with at least a few touches.