logical phenomena as will enable us to rise to the supreme position of being able to predict the facts of climate with assured accuracy, and for a long time in advance. Let us first enumerate some of the particular phenomena which are necessarily more or less connected together. The most fundamental of all the elements concerned is the pressure of the air as indicated by the barometer; then there is the temperature of the air and the degree of its saturation, the amount and character of the clouds, the rainfall, together with comparatively exceptional incidents such as hailstorms and thunderstorms. At present, no doubt, we are enabled, by the careful collection of observations all over the world, to predict in some degree the recurrence of these phenomena. Our newspapers give us each morning a forecast of the kind of weather that may be expected. But every one knows that, though these forecasts are often useful, they yet have a very inferior degree of accuracy to the kind of prediction which we find in the "Nautical Almanac," where the occurrence of an eclipse of the moon, or of an occultation of a star, or a transit of Venus, or any similar astronomical event, is foretold with definiteness and with perfect certainty of fulfilment. Yet no one can really doubt that the temperature at London Bridge next Christmas Day, or the height of the barometer at noon on January 1, 1900, are each of them quite as certainly decided by law as the time of high water or any other astronomical element.
We know that there will be a transit of Venus in the year A.D. 2004, and that there will be no such phenomenon until then, while there will be a repetition of the occurrence in A.D. 2012. It is certain that these predictions will be fulfilled, yet why is it that we can
