Or to take an instance from Astronomy. We are accustomed to say that the path of the moon is an ellipse. But this is a very loose statement. On enquiry we find that, owing to perturbations said to be produced by the sun, the path deviates considerably from an ellipse. In fact in strict calculations it is taken as being a certain ellipse only for an instant--the next instant it is supposed to be a portion of another ellipse. We might then call the path an irregular curve somewhat resembling an ellipse. This is a new view. But on further enquiry it appears that, while the moon is going round the earth, the earth itself is speeding on through space about the sun--in consequence of which the actual path of the moon does not in the least resemble an ellipse! Finally the sun itself is in motion with regard to the fixed stars, and _they_ are in movement too. What then is the path of the moon? No one knows; we have not the faintest idea--the word itself ceases to have any assignable meaning. It is true that if we agree to ignore the perturbations produced by the sun--as in fact we _do_ ignore perturbations produced by the planets and other bodies--and if we agree to ignore the motion of the earth, and the flight of the solar system through space, and even the movement of any centre round which that may be speeding, we may then _say_ that the moon moves in an ellipse. But this has obviously nothing to do with actual facts. The moon _does_ not move in an ellipse--not even "relatively to the earth"--and probably never has done and never will do so. It may be a convenient view or fiction to say that it would do so under such and such circumstances--but it is still only a fiction. To attempt to isolate a small portion of the phenomena from the rest in a universe of which the _unity_ is one of Science's most cherished convictions, is obviously self-stultifying and useless.