of the activity of these forces from the remark of M. Arago, that frequently on the same days, when he had observed at Paris violent disturbances of the regular movement of the magnetic needle, northern lights, not visible above the horizon of Paris, had been seen at distant places.
The irregularities in the phænomena of terrestrial magnetism, the frequent occurrence of which had also been observed, especially by Humboldt in his numerous observations of the diurnal and horary oscillations of the magnetic needle, thus obtained a peculiar interest. Though the facts which had been remarked, neither proved that all irregular oscillations of the needle are contemporaneous with the northern lights, nor precluded the possibility, that many, perhaps most of them, have merely local causes, yet it was scarcely possible to mistake the evidence of the not unfrequent action, over a wide extent, of great natural forces, which, if they could not yet be investigated in their sources, offered at least a worthy object of natural inquiry, in respect to the relations of their activity and extent.
Superficial and merely accidental recognitions of such relations can bring us no nearer to this goal: in order to attain it, many such phænomena must be contemporaneously followed up in accurate detail at numerous stations, and their time and magnitude closely ascertained and measured. For this purpose, however, previously concerted plans are essentially necessary among those observers who have suitable means at their disposal.
The celebrated philosopher to whom we are indebted for so many additions to our knowledge of terrestrial magnetism was also here the first to lead the way. M. von Humboldt caused to be erected in Berlin, towards the end of the year 1828, a small house, free from iron,—placed in it a variation compass constructed by Gambey,—and concerted with possessors of similar instruments at various places, some of which were very distant, regular observations of the magnetic variation on fixed days. Eight terms in the year were agreed upon, each of forty-four hours, dui'ing which the variation was to be noted from hour to hour; at some places observations were made within still narrower limits of time, viz. at every half-hour, or every twenty minutes. The details will be found in the nineteenth volume of Poggendorff's Annalen der Physik, p. 361; and in the same journal are also the observations which, according to this agree-
