Strange to say, one of them is revolving about the other in a little curve. The pole of figure is travelling around the pole of rotation. The distance between them varies a little, never becoming greater than about fifty feet, and it takes about fourteen months to complete a revolution. There are some slight irregularities in the motion, but, in the main, it takes place in the manner here stated. In consequence of this rotation of the one pole about the other, the pole of figure is now on one side of the rotation pole and now on the opposite side, but it never travels continuously in one direction. Thus, as we have already seen, the sort of continuous motion required to explain the observed geological phenomena has not yet been found by astronomers. Observations for the study of latitude variations have been made very extensively within recent years both in Europe and the United States. It has been found practically most advantageous to carry out simultaneous series of observations at two observatories situated in widely different parts of the earth, but having very nearly the same latitude. It is then possible to employ the same stars for observation in both places, whereas it would be necessary to use different sets of stars if there were much difference in the latitudes. There is a special advantage in using the same stars in both places. We can then determine the small difference in latitude between the two participating observatories in a manner which will make it quite free from any uncertainty in our knowledge of the positions on the sky of the stars observed; for, strange as it may seem, our star-catalogues do not contain absolutely accurate numbers. Like all other data depending on fallible human observation, they are affected with small errors. But if we can determine simply the difference in latitude of the two observatories, we can discover from its variation the path in which the pole is moving. If, for instance, the observatories are separated by one-quarter the circumference of the globe, the pole will be moving directly toward one of them, when it is not changing its distance from the other one at all. This method was used for seven years with good effect at the observatories of Columbia University in New York, and the Royal Observatory at Naples, Italy. For obtaining its most complete advantages it is, of course, better to establish several observing stations on about the same parallel of latitude. This was done in 1899 by the International Geodetic Association. Two stations are in the United States, one in Japan, and one in Sicily. We can, therefore, hope confidently that our knowledge as to the puzzling problem of polar motion will soon receive very material advancement. SATURN'S RINGS