No theory of the origin of comets ever suggested, none even which can be imagined, could account for such a peculiarity. Whereas, on the other hand, we have direct evidence showing how a comet, originally single, may be transformed into two or more comets traveling on the same, or nearly the same, track. The comet called Biela's, which had circuited as a single comet up to the year 1846 (during a period of unknown duration in the past--probably during millions of years), divided then into two, and has since broken up into so many parts that each cometic fragment is separately undiscernible. The two comets into which Biela's divided, in 1846, were watched long enough to show that had their separate existence continued (visibly), they would have been found, in the fullness of time, traveling at distances very far apart, though on nearly the same orbit. The distance between them, which in 1846 had increased only to about a quarter of a million of miles, had in 1852 increased to five times that space. Probably a few thousands of years would have sufficed to set these comets so far apart (owing to some slight difference of velocity, initiated at the moment of their separation) that when one would have been at its nearest to the sun, the other would have been at its farthest from him. If we could now discern the separate fragments of the comet, we should doubtless recognize a process in progress by which, in the course of many centuries, the separate cometic bodies will be disseminated all round the common orbit. We know, further, that already such a process has been at work on portions removed from the comet many centuries ago, for as our earth passes through the track of this comet she encounters millions of meteoric bodies which are traveling in the comet's orbit, and once formed part of the substance of a comet doubtless much more distinguished in appearance than Biela's. There can be little doubt that this is the true explanation of the origin of that family of comets, five of whose members returned to the neighborhood of the sun (possibly their parent) in the years 1668, 1843, 1880, 1882, and 1887.[1] [Footnote 1: It may be interesting to compare the orbital elements of the five comets above dealt with. They may be presented as follows; but it should be noticed that the determinations must be regarded as rough in the case of Comets I. and V., as the observations were insufficient for exact determination of the elements: