All northern navigators and travellers agree, that high north the sun becomes less bright, and the sky darker, than in more southern latitudes. Is not this owing to the rays of the sun being refracted round the verge of the polar opening? Another circumstance, observed by navigators, who have visited high latitudes is, that the latitude and longitude, as found by celestial observation, frequently differ very materially, sometimes as much as one half, from that given by the log-line.[28] It has also been observed that the mercury in the barometer is less fluctuating in northern regions, than it is further south. Those appearances observed in the southern hemisphere, which are termed Magellanic clouds, by navigators, have not, so far as I know, been accounted for. They are three in number, of an irregular shape, and observed by night in the South Atlantic, and the south-east parts of the Pacific oceans, (reversed from New-Holland and New-Zealand,) but never visible in the eastern parts of the Indian ocean: their colour is like that of far distant mountains, on which the sun is shining. In the one sea they appear due south, and in the other to the left. They are stationary, appearing perpetually fixed at a certain height, and in a particular situation, as viewed from any given place. The stars and the heavens, in their diurnal revolutions, sweep by them, and they remain the same. To the navigator, who proceeds to the east or west, they appear to be more or less to the right or left of the meridian, in proportion as he changes his longitude; and as he sails south, they increase in height, until they reach the zenith, and finally become north, when seen by an observer south of the straits of Magellan, which is in latitude fifty-two degrees south. Captain Symmes accounts for the appearance of these clouds by the great refractive power of the atmosphere about the polar openings; causing the opposite side of the verge to appear pictured in the sky, as navigators inform us objects do sometimes appear, in the arctic regions; and in the manner Scoresby's ship appeared in the sky, with every particular about her so accurately represented, as to be at once identified by the observers, though the vessel, at that time, was at such a distance as to render it rather incredible how she could be seen at all. As proof of this position, Captain Symmes alleges, that the relative position, shape, and proportions of these clouds, agree in their general outlines with the southern part of New-Zealand, the southeast part of New-Holland, and the whole of Van-Dieman's land, which are situated on, and near to the verge of the sphere, opposite to where the clouds are visible. These clouds are only seen in the night when the atmosphere is clear, at which time the sun is shining on the islands in question. Hence it is alleged, that from these facts, their relative appearance is deducible. As we are never sensible that the rays of light are refracted by the