A further argument in favour of the meteoric origin of solar alternations of heat is to be found in the fact that, on one occasion at least, a solar phenomenon, corresponding precisely to what we should expect to see, if great meteoric masses fell upon the sun, has been followed by precisely the same signs of terrestrial disturbance which accompany and follow the formation of great solar spots. I refer to the remarkable occurrence witnessed by Carrington and Hodgson (at different observatories) in September, 1859, when two intensely bright points of light were seen travelling beside each other at the rate of about 120 miles per second along a short arc of the sun's surface,--an arc only equal in length to some four-and-a-half times the diameter of our earth. On that occasion the emission of solar heat may or may not have been increased in an appreciable degree for several minutes. My own belief is that it must have been; but we certainly have no means of proving that it was. What we do know certainly is, that on that day all the phenomena which usually accompany the existence of many and large sun-spots showed themselves with exaggerated intensity. The magnetic needle was greatly disturbed, auroras displayed their coloured streamers in both hemispheres, telegraphic communication was interrupted, and everything tended to show that a disturbance of the same general character as that which produces sun-spots, but much more active while it lasted, had affected the sun. It seems, then, altogether reasonable to infer that sun-spots are due to the same cause as the disturbance which then occurred. So that if we conclude, with most astronomers competent to form an opinion, that the disturbance witnessed by Carrington and Hodgson was due to the downfall of two very large meteoric masses upon the sun, it would follow that sun-spots are due to more wide-spread meteoric showers, not consisting of masses so large. The reader will long since have guessed, no doubt, to what all this tends. If the periodical variations of the sun's surface are due to meteoric and cometic systems whose orbits intersect the sun's globe, their periods being short (that is, lasting but a few years), it may well be that more important meteoric and cometic systems intersecting the sun's globe exist, which have much longer periods. When next one of these makes its passage athwart the sun, far more important solar disturbances may take place than those which occur when the regularly recurring systems salute the sun. Two or three times in the history of science comets have approached very close to the surface of the sun, as in 1680, and again in 1843, but without actually impinging upon it. Very slight changes in the motions of those comets, owing to the disturbing influences of the planets, would cause their very nuclei to strike the sun, and their meteoric trains to pour afterwards in a full stream upon him for many days, or even for many months and years in succession.