9. In close connection with this point, I may refer to the relation between the weather generally which prevails from time to time upon the earth, and the periods of sparse or copious spots. The conditions which we comprehensively term "weather" are obviously dependent upon the energy, abundant or diminished, which is communicated from the sun in the modes of light and heat; and it is reasonable to infer that great changes in the one will be associated with similar ranges of modification in the other. But observed facts must be the staple of our inductions, and facts are not yet fully decisive; it is admitted, however, on such evidence as has been procurable, that some physical connection exists between spot-cycles and the periodical variations in weather, such as hot and cold years, wet and dry years, years of prolific and deficient harvests; but the direct relationship is so masked by other causes, which act in the production of weather, that any attempt at precise definition of the actual effect of solar maculation is not yet practicable. Mr. J. A. Broun, the well-known meteorologist, informed Professor Balfour Stewart that in his judgment all "meteorological phenomena are due to solar action; that the heating action of the sun is not the only one; but that the action which produces variations in the earth' s magnetic force affects the conditions of the atmospheric gases, introducing forces which we cannot, in the present state of our knowledge, appreciate, though the facts appear to me to prove their existence."

10. Dr. A. Schuster has shown, from an investigation into the relevant periods, that what have been considered to be good wine years in Germany have closely corresponded with minima of spots. Thus the years when spots were least prevalent were 1810.5 (that is, midway between 1810 and 1811), 1823.2, 1833.8, 1844, 1856.2 and 18672, while the coincident years of excellent vintages were 1811, 1822, 1834, 1846, 1857-8 and 1868. Since the quality of the grape depends upon abundant sunshine, it is observed that this condition is fulfilled during a minimum period of spots, when the entire area of the photosphere which constitutes the source of heat is the least riven with the rents and cavities which sun spots mean.

11. The inquiry is most interesting whether any causal relation subsists between a prosperous or adverse condition of harvests and crops (especially of wheat) and the frequency or infrequency of spots. The original and penetrating mind of Sir W. Herschel arrived at the conclusion that years of diminished spots coincided with years when the price of corn was low, while the price was increased when the area of spots became augmented, so that the fewer the spots, he concluded, the more favourable was the influence of the solar rays upon the growth of crops. This appears reasonable (though actual observations should be substituted for a priori considerations) since the years of least numerous spots are years of more continuous and equable sunshine and heat in consequence of the photosphere then possessing a uniformly unrent surface. This connection, if established, would concur with the statement relating to superior vintage years.

12. This suggestion naturally leads to the important investigations of Professor Stanley Jevons into the relation between commercial crises - one of the phases presented uniformly by the course of trade - and the sun's condition as manifested in the form of periodic spots. From an exhaustive examination of records Professor Jevons has found that the chief commercial crises or financial collapses (confining our survey to the nineteenth century and to the close of Jevons's writings) occurred in 1825, 1836-9 (1837 in the United States), 1847, 1867 and 1866, showing an average interval of 10.8 years, which is closely coincident with the period from maximum to maximum, and minimum to minimum, of the solar spots. And including the crises which appeared in the eighteenth century and in 1804-5, 1815 and 1878 (the latter of which Professor Jevons predicted), he obtained an average period of 10.466 years, which virtually accords with Mr. Broun's result (the most recent and exact) of 10.45 for the duration of a solar cycle. Now the theory of probability (the sole guide in inquiry and conduct) shows that two periodic phenomena (spots and collapses of trade), varying in virtual coincidence, are causally connected. As Professor Foxwell has remarked, the conclusion which Jevons claimed to have reasonably and scientifically established was that the periodic variations in harvests are associated with the solar period, and that this harvest variation operates in the stimulation and determination of the naturally rhythmic fluctuations of European trade. For, as Jevons himself states, the fluctuations in the money market show a remarkable tendency to recur at intervals approximating very nearly to the range between maxima and maxima and minima and minima of spots. If, he says, the state of the sun governs vintages and harvests, it thereby determines the prices of food and raw materials, and as these constitute the articles with which commerce deals, the condition of the money market is also involved, since the latter depends upon the relation between the supply and demand of those commodities. These (practically) decennial crises depend upon meteorological variations of similar period, and these again depend upon cosmical variations of whose existence we possess evidence in the frequency of spots, of auroras, and magnetic perturbations. It would appear clear that Professor Jevons's conclusion will be converted into a scientific proposition with further and more exhaustive research - the essential point being that the two cycles of trade (from soundness through inflation to collapse) and of solar commotion are in exact parallelism - both as to course and boundaries - to each other. One or two coincidences between changes in terrestrial phenomena and the modifications in solar activity might be due to chance - that is, to forces of which we possess no knowledge or conception. But when we compare the different and concurrent results which this exposition has presented we are compelled to conclude a definite relation of cause and effect in accordance with an important and established canon of inductive inference - that "whatever phenomenon varies in any "manner whenever another phenomenon varies in some "particular manner, is either a cause or an effect of that "phenomenon, or is connected with it through some fact "of causation." (Mill's System of Logic, Vol. I, Book III, cap. viii, § 6.)