When springs charged with silica or carbonate of lime rise upon the slope of a hill composed of loose volcanic materials, they give rise to the remarkable structures known as sinter- and travertine-terraces (see fig. 80). The water flowing downwards from the vent forms a hard deposit upon the lower slope of the hill, while the continual deposition of solid materials within the vent tends to choke it up. As a new vent cannot be forced by the waters through the hard rock formed below, it is originated a little higher up. Thus the site of the spring is gradually shifted farther and farther back into the hill. As deposition takes place along the surfaces over which this water flows, terraces are built up enclosing basins. Of structures of this kind we have remarkable examples in the sinter-terraces of Rotomahana in New Zealand and the travertine-terraces of the Gardiner's River in the Yellowstone Park district of the Rocky Mountains. [Illustration: Fig. 80.--Diagram illustrating the mode of formation of Travertine and Sinter Terraces on the sides of a hill of tuff.] CHAPTER VII. THE SUCCESSION OF OPERATIONS TAKING PLACE AT VOLCANIC CENTRES. That a volcanic vent, when once established, may display intense activity during enormous periods of time, there cannot be the smallest reason for doubting; for the accumulation of materials around some existing volcanic centres must certainly have been going on during many thousands, perhaps millions, of years. To us, whose periods of observation are so circumscribed, it may therefore at first sight appear a hopeless task to trace the 'life-history of a volcano,' to discover the stages of its development, and to indicate the various episodes which have occurred during the long periods it has been in existence. But when it is remembered that we have the opportunity of studying and comparing hundreds of such volcanoes, exhibiting every varying phase of their development, we shall see that such an attempt is by no means so unpromising as it at first sight appears to be. In the present chapter, we shall give an account of the results which have already been obtained by inquiries directed to this object. [Sidenote: CYCLES OF VOLCANIC PHENOMENA.] There is not the smallest room for doubt that during the past history of our globe, exhibitions of subterranean energy have occurred at many different parts of its surface. There is further evidence that at the several sites where these displays of the volcanic forces have taken place, the succession of the outbursts has run through a regular cycle, gradually increasing in intensity to a maximum, and then as gradually dying away.