We may carry the canal idea a step further, with a remarkably interesting result. Suppose the whole surface of the earth were covered with a series of canals parallel to the equatorial canal. Then, as we approach the pole, the canals will be shorter, since the equatorial circumference of the earth is longer than it is in any other latitude. Thus the waves in high latitudes would not have so far to go as the waves in low latitudes, and so might tend to reinforce each other as ex plained above. So we might have 'inverted' tides in the equatorial regions and direct ones in the polar regions; and in some intermediate latitude there would be, as Darwin says, "very great tides, the nature of which cannot be speci fied exactly." This would occur, as we have seen, where the earth's circumference is short enough to permit a free wave to go all around in about twenty-four hours. These dynamical considerations of wave motion in canals lead to results bearing some sort of resemblance to the actual observed phenomena of nature. It should be noted, of eourse, that the foregoing considera tions refer to theoretical conditions such as would exist if the earth were simply covered with a layer of water. But the actual facts of nature are so different that they modify very greatly the theoretical tidal action of the sun and moon. So large a part of the terrestrial sur
face is covered by land that the free motion of tidal waves is seriously impelled. It is therefore impossible to predict the times of high water aceurately from theoretical considerations alone. Fortunately, the practical prediction of the time of high water for any place call be effected by analyzing a long series of tidal observations made at the place in question. This method of pro cedure has been in use for many years, and we now possess tidal tables for all principal sea ports accurate enough for the purposes of navi gation.
An interesting modification of ordinary tides occurs in rivers. Here the tidal rise appears of course as a strong current running up-stream, where the ocean level outside the river-mouth has been raised. At times this tidal current advances with a high and dangerous wave crest (called a borc), and it may acquire velocity enough to raise the water-level in the river con siderably above that of the ocean outside. Phe nomena of this kind, and indeed all tidal phe nomena, are partly modified by the configuration of the coast line and the depth of water.