This phenomenon is absolutely general and is observed on the entire globe, from the equator to the poles, in the same manner; the amplitude of the oscillation increases with the latitude but not proportionately; it only amounts to 1 or 2 minutes of arc in the tropics, to 9 minutes in France, and 7 minutes in Norway. This variation corresponds sensibly to the variation of temperature, the amplitude of which increases from the tropics to the poles. Heat, electricity, water-vapour, and barometric pressure are all associated with it. There are certain perturbations, to be mentioned presently, to which some human beings are sometimes a little too sensive. We live enveloped in an invisible world. The amplitude of this daily oscillation varies from day to day, from month to month, from year to year. If we take the mean of the observations of a whole year, we find that from one year to another it changes from a certain amount to double that amount and that this annual variation is regulated by a law. It is periodic, and the average period is eleven or twelve years. For instance, in 1870 the oscillation was twelve minutes of arc and in 1878 it was six minutes. I take the figures for London, Prague, Munich, Rome, and Milan--not those for France, for no observations were made in that terrible year, or they were made badly, and I made many enemies by insisting upon them. That year, 1870, was a maximum. There were other maxima, but not quite so great, in 1883, 1893, 1905, and in 1916. The cycle varies from ten to thirteen years. Now, this behaviour of the magnetic needle corresponds exactly to the state of health of the Sun, i.e. its activity or the number of sunspots or whirlpools which indicate it as well as its gigantic flames, protuberances, faculæ, and the various manifestations of its radiating power. There was a very considerable maximum of sunspots in 1870. Another maximum occurred in 1883 and another in 1905, but the latter were less marked than that of 1870. In 1918 another very well defined maximum occurred. Minima of sunspots, protuberances, and faculæ were observed in 1878, 1889, 1901, and 1913. Now, these maxima and minima correspond exactly with the variation of the compass needle, as is shown in the tables and curves published in my _Astronomie Populaire_. If curves are traced for solar activity on the one hand and magnetic oscillation on the other hand, it is found that the two curves are absolutely parallel. Science has been able to trace this parallelism very far, as far back indeed as the eighteenth century, and it is found to include the aurora borealis, which is a magnetic phenomenon. The concordance is evident and absolutely incontestable. Thus there constantly emanates from the Sun a force different from light and heat, a force we do not perceive with our senses and which places our small and mobile planet in constant communication with our central star, which is more than a million times greater in volume.