Comets move in every direction in their orbits, and these orbits have every conceivable inclination to the ecliptic. 292. _Periodic Comets._--There are quite a number of comets which are known to be _periodic_, returning to the sun at regular intervals in elliptic orbits. Some of these have been observed at several returns, so that their period has been determined with great certainty. In the case of others the periodicity is inferred from the fact that the velocity fell so far short of the parabolic limit that the comet must move in an ellipse. The number of known periodic comets is increasing every year, three having been added to the list in 1881. The velocity of most comets is so near the parabolic limit that it is not possible to decide, from observations, whether it falls short of it, or exceeds it. In the case of a few comets the observations indicate a minute excess of velocity; but this cannot be confidently asserted. It is not, therefore, absolutely certain that any known comet revolves in a hyperbolic orbit; and thus it is possible that all comets belong to our system, and will ultimately return to it. It is, however, certain, that, in the majority of cases, the return will be delayed for many centuries, and perhaps for many thousand years. 293. _Origin of Comets._--It is now generally believed that the original home of the comets is in the stellar spaces outside of our solar system, and that they are drawn towards the sun, one by one, in the long lapse of ages. Were the sun unaccompanied by planets, or were the planets immovable, a comet thus drawn in would whirl around the sun in a parabolic orbit, and leave it again never to return, unless its path were again deflected by its approach to some star. But, when a comet is moving in a parabola, the slightest _retardation_ would change its orbit to an ellipse, and the slightest _acceleration_ into a hyperbola. Owing to the motion of the several planets in their orbits, the velocity of a comet would be changed on passing each of them. Whether its velocity would be accelerated or retarded, would depend upon the way in which it passed. Were the comet accelerated by the action of the planets, on its passage through our system, more than it was retarded by them, it would leave the system with a more than parabolic orbit, and would therefore move in a hyperbola. Were it, on the contrary, retarded more than accelerated by the action of the planets, its velocity would be reduced, so that the comet would move in a more or less elongated ellipse, and thus become a permanent member of the solar system.