Let us now examine another phenomenon. The Aurora Borealis has been generally considered to be in some way connected with the magnetism of the earth, and with the position of the magnetic pole. It is certain that the appearance of this meteor does affect the needle in a way not to be mistaken, and (although not invariably) the vertex of the luminous arch will usually conform to the magnetic meridian. Yet (and this is worthy of attention), the observations made in the North Polar Expeditions[29] "appear to prove, that in the immediate vicinity of the magnetic pole the development of light is not in the least degree more intense or frequent than at some distance from it." In fact, as the American magnetic pole is, as stated, in latitude 73°, the central vortex will seldom reach so high, and, consequently, the aurora ought at such times to be more frequent in a lower latitude. In a late work by M. de la Rive, this gentleman expresses the opinion, that the cause of the aurora is not due to a radiation of polar magnetism, but to a purely electrical action.[30] His explanation, however, is not so satisfactory as his opinion. Now, we have examined numerous cases of auroral displays, and never yet found one which could not be legitimately referred to the action of ethereal vortices. Generally, the aurora will not be visible, when the upper surface of the atmosphere of that latitude in which the vortex is known to be (reckoning in the direction of the magnetic meridian) is below the horizon, which shows that the brightest portion is in the atmosphere. In latitude 41° even, it may show itself when the vortex is three days north, more frequently when one or two days north; but when the vortex passes centrally, or south, it rarely is seen, and this is the only difficulty in explaining it by the theory. But, when we reflect that the ether shoots out in straight lines, and at an angle corresponding to the magnetic dip, we are at no loss to perceive the reason of this. If each minute line composing the light were seen endwise, it would be invisible; if there were millions such in the same position, they could add nothing to the general effect; but, when viewed sideways, the case would be different, there would be a continued reduplication of ray upon ray, until in the range of some hundreds of miles an effect might be produced amounting to any degree of intensity on record. Now, this is the case when the aurora is immediately overhead, it will be invisible to those below, but may be seen by persons a hundred miles south; so, also, when it is to the south, it is too oblique to the line of vision to be seen, especially as all the rays to the northward of the observer can contribute nothing to increase the effect. That it is of the nature of rays very much diffused, can hardly be doubted; and, therefore, if only of a few miles in depth, its impressions are too faint to be sensible. By referring to