predictions have been fulfilled. The objection I refer to is this: we know that the weather is seldom alike over very wide regions, while, nevertheless, the celestial bodies present the same aspect toward the whole extent of such regions, or an aspect so nearly the same as to suggest that the same conditions of weather should prevail if the weather really depended on the position of the heavenly bodies. It appears, then, that the inventor of a really trustworthy system must have a distinct scheme for each part of every continent—nay, of every country, if not of every county. This objection is not taken into account, however, by the inventors of systems, while the fact on which it depends affords the means of showing that each prediction has been fulfilled. Thus, suppose "bad weather and much wind" have been predicted on a certain day, and that day is particularly fine and calm in London. If this were urged as an objection to the soundness of the system, the answer would run somewhat on this wise: "Unquestionably it was fine in London, but in North Scotland (or in France, or Spain, or Italy, as the case may be) there was very gloomy weather, and in Ireland (suppose) quite strong winds are reported to have prevailed in the afternoon." The readiness with which men satisfy themselves in such cases, corresponds with that mischievous ingenuity wherewith foolish persons satisfy themselves that a fortune-teller had foretold the truth, that a dream had been fulfilled, a superstition justified, and so forth.
The tendency, at present, among those who are desirous of forming a scheme of weather prediction, is to seek the origin of our weather-changes in changes of the sun's condition, and, by determining the laws of the solar changes, to ascertain the laws which regulate changes in the weather.
It may be remarked, in passing, that this new phase of the inquiry does not reject planetary influences altogether. The theory is entertained by many well-known students of science that changes in the condition of the sun are dependent on the varying positions of the planets; so that, if it should be established that our weather-changes are connected with solar changes, we should infer that indirectly the planets in their motions rule the weather on our earth.
I propose now to consider the evidence relating to the sun's influence, and to discuss the question (altogether distinct, be it remarked) whether a means of accurate weather prediction may be obtained if the sun's influence be regarded as demonstrated.
There is one strong point in favor of the new theory, in the fact that the sun is unquestionably the prime cause of all weather changes. To quote the words of Lieutenant-Colonel Strange, an enthusiastic advocate of the theory (and eager to have it tested at this country's charge), "there can hardly be a doubt that almost every natural phenomenon, connected with the climate can be distinctly traced to the sun as the great dominating force, and it is a natural inference" (though not, as
