A comparison of the facts known in regard to comets, falling-stars, and meteoric stones, seems to warrant the inference that they are bodies of the same nature, and perhaps of similar origin; differing from each other mainly in the accidents of magnitude and density. The hypothesis of Laplace very obviously accounts for the formation of planets and satellites, moving in the same direction, and in orbits nearly circular; but how, it may be asked, can the same theory explain the extremely eccentric, and in some cases retrograde, motions of comets and aerolites? This is the question to which we now direct our attention. After the nuclei of the solar and sidereal systems had been established in the primitive nebula, and when, in consequence, immense gaseous spheroids had collected around such nuclei, we may suppose that about the points of equal attraction between the sun and neighboring systems, portions of nebulous matter would be left in equilibrio. Such outstanding nebulosities would gradually contract through the operation of gravity; and if, as would sometimes be the case, the solar attraction should preponderate, they would commence falling toward our system. Unless disturbed by the planets they would probably move round the sun in parabolas. Should they pass, however, near any of the large bodies of the system, their orbits might be changed into ellipses by planetary perturbation. Such was the view of Laplace in regard to the origin of comets. It seems probable, however, that many of these bodies originated _within_ the solar system, and belong properly to it. The outer rings thrown off by the planets may have been at too great distances from the primaries to form stable satellites. Such masses would be separated by perturbation from their respective primaries, and would revolve round the sun in independent orbits. Again: small portions of nebulous matter may have been abandoned as primary rings, at various intervals between the planetary orbits. At particular distances such rings would be liable to extraordinary perturbations, in consequence of which their orbits would ultimately assume an extremely elliptical form, like those of comets, and perhaps also those of meteors. It was shown in Chapter XIII. that several such regions occur in the asteroid zone between Mars and Jupiter. We may add, in confirmation of this view, that there are twelve known comets whose periods are included between those of Flora and Jupiter. Their motions are all direct; their orbits are less eccentric than those of other comets; and the mean of their inclinations is about the same as that of the asteroids. These facts certainly appear to indicate some original connection between these bodies and the zone of minor planets.