sun, or it might of course have been opposite thereto.  As a matter of fact the two agree.  The moon in its monthly revolution around the earth follows also the same direction, and our satellite rotates on its axis in the same period as its monthly revolution, but in doing so is again observing this same law.  We have therefore in the earth and moon four movements, all taking place in the same direction, and this is also identical with that in which the sun rotates once every twenty-five days.  Such a coincidence would be very unlikely unless there were some physical reason for it.  Just as unlikely would it be that in tossing a coin five heads or five tails should follow each other consecutively.  If we toss a coin five times the chances that it will turn up all heads or all tails is but a small one.  The probability of such an event is only one-sixteenth.

There are, however, in the solar system many other bodies besides the three just mentioned which are animated by this common movement.  Among them are, of course, the great planets, Jupiter, Saturn, Mars, Venus, and Mercury, and the satellites which attend on these planets.  All these planets rotate on their axes in the same direction as they revolve around the sun, and all their satellites revolve also in the same way.  Confining our attention merely to the earth, the sun, and the five great planets with which Laplace was acquainted, we have no fewer than six motions of revolution and seven motions of rotation, for in the latter we include the rotation of the sun.  We have also sixteen satellites of the planets mentioned whose revolutions round their primaries are in the same direction.  The rotation of the moon on its axis may also be reckoned, but as to the rotations of the satellites of the other planets we cannot speak with any confidence, as they are too far off to be observed with the necessary accuracy.  We have thus thirty circular movements in the solar system connected with the sun and moon and those great planets than which no others were known in the days of Laplace.  The significant fact is that all these thirty movements take place in the same direction.  That this should be the case without some physical reason would be just as unlikely as that in tossing a coin thirty times it should turn up all heads or all tails every time without exception.

We can express the argument numerically.  Calculation proves that such an event would not generally happen oftener than once out of five hundred millions of trials.  To a philosopher of Laplace’s penetration, who had made a special study of the theory of probabilities, it seemed well-nigh inconceivable that there should have been such unanimity in the celestial movements, unless there had been some adequate reason to account for it.  We might, indeed, add that if we were to include all the objects which are now known to belong to the solar system, the argument from probability might be enormously increased in