“But can we predict?” We certainly cannot till we have, at any rate, first shown that there is a connection between sun-spots and the weather. Since we know nothing of the ultimate causes involved, we can only at present, as I say, collect records of the changes there, and compare them with others of the changes here, to see if there is any significant coincidence. To avoid columns of figures, and yet to enable the reader to judge for himself in some degree of the evidence, I will give the results of some of these records represented graphically by curves, like those which he may perhaps remember to have seen used to show the fluctuations in the value of gold and grain, or of stocks in the stock-market. It is only fair to say that mathematicians used this method long before it was ever heard of by business men, and that the stockbrokers borrowed it from the astronomers, and not the astronomers from them. In Fig. 48, from Carrington’s work, each horizontal space represents ten years of time, and the figures in the upper part represent the fluctuations of the sun-spot curve. In the middle curve, variations in vertical distances correspond to differences in the distance from the sun of the planet Jupiter, the possibility of whose influence on sun-spot periods can thus be examined. In the third and lowest, suggested by Sir William Herschel, the figures at the side are proportional to the price of wheat in the English market, rising when wheat ruled high, falling when it was cheap. In all three curves one-tenth of a horizontal spacing along the top or bottom corresponds to one year; and in this way we have at a glance the condensed result of observations and statistics for sixty years, which otherwise stated would fill volumes. The result is instructive in more ways than one. The variations of Jupiter’s distance certainly do present a striking coincidence with the changes in spot frequency, and this may indicate a real connection between the phenomena; but before we decide that it does so, we must remember that the number of cycles of change presented by the possible combination of planetary periods is all but infinite. Thus we might safely undertake, with study enough, to find a curve, depending solely on certain planetary configurations, which yet would represent with quite striking agreement for a time the rise and fall in any given railroad stock, the relative numbers of Democratic and Republican congressmen from year to year, or anything else with which the heavenly bodies have in reality as little to do. The third curve (meant by the price of wheat to test the possible influence of sun-spots on years of good or bad harvests) is not open to the last objection, but involves a fallacy of another kind. In fact the price of wheat depends on many things quite apart from the operations of Nature,--on wars and legislation, for instance; and here the great rise in the first years of the century is as clearly connected with the