THE “CAPTURE THEORY” OF COMETS.—A memoir by M. O. Callandreau, on the capture theory of periodic comets, has recently been published (Annales de l'Observatoire de Paris , vol. xx.). It is generally known that the periodic comets are distributed in groups which depend in some manner on the major planets. Jupiter's family of comets is at least fifteen in number, and all the members of it have direct motion, orbits only slightly inclined to the orbit of Jupiter, and aphelion points near Jupiter's aphelion; what is more—one of the two points where each of them intersects the plane of Jupiter's orbit is generally very near to the trajectory of this planet. The theory which best explains such distribution is that which regards the comets of which the groups are composed as having come under the perturbing influence of the major planet to which they are re spectively related. If a comet arrives from interstellar space into the solar system with a sensible parabolic velocity, and passes near a major planet, the velocity will be either diminished or increased. In the former case, the parabolic orbit would be transformed into an elliptical one, and the comet would be, as it were, incorporated into our system-captured by the planet. If, on the other hand, the velocity is accelerated, the orbit becomes hyperbolic, and the comet moves away from our system, never to return. The results of a research on this subject were given by M. Tisserand a few months ago (Bulletin Astronomique, July 1889, and NATURE, vol. xlii. p. 31).