either by its conformity to known or ascertainable facts—as when Kepler determined the elliptic orbit of Mars; or by the fulfilment of the predictions it has sanctioned—as in the occurrence of an eclipse or an occultation at the precise moment specified many years previously; or, still more emphatically, by the actual discovery of phenomena till then unrecognized—as when the perturbations of the planets, shown by Newton to be the necessary results of their mutual attraction, were proved by observation to have a real existence; or, as when the unknown disturber of Uranus was found in the place assigned to him by the computations of Adams and Le Verrier.
We are accustomed, and I think most rightly, to speak of these achievements as triumphs of the human intellect. But the very phrase implies that the work is done by mental agency, and the coincidence of its results with the facts of observation is far from proving the intellectual process to have been correct. For we learn, from the confessions of Kepler, that he was led to the discovery of the elliptic orbit of Mars by a series of happy accidents, which turned his erroneous guesses into the right direction; and to that of the passage of the radius vector over equal areas in equal times by the motion of a whirling force emanating from the sun, which we now regard as an entirely wrong conception of the cause of orbital revolution. It should always be remembered, moreover, that the Ptolemaic system of astronomy, with all its cumbrous ideal mechanism, did intellectually represent all that the astronomer, prior to the invention of the telescope, could see from his stand-point the earth with an accuracy which was proved by the fulfilment of his anticipations. And in that last and most memorable prediction, which has given an imperishable fame to our two illustrious contemporaries, the inadequacy of the basis afforded by actual observation of the perturbations of Uranus required that it should be supplemented by an assumption of the probable distance of the disturbing planet beyond, which has been shown by subsequent observation to have been only an approximation to the truth.
Even in this most exact of sciences, therefore, we cannot proceed a step without translating the actual phenomena of Nature into intellectual representations of those phenomena, and it is because the Newtonian conception is not only the most simple, but is also, up to the extent of our present knowledge, universal in its conformity to the facts of observation, that we accept it as the only scheme of the universe yet promulgated which satisfies our intellectual requirements.
When, under the reign of the Ptolemaic system, any new inequality was discovered in the motion of a planet, a new wheel had to be added to the ideal mechanism—as Ptolemy said, "to save appearances." If it should prove, a century hence, that the motion of Neptune himself is disturbed by some other attraction than that exerted by the interior planets, we should confidently expect that not an ideal but a real cause for that disturbance will be found in the existence of
