On September 17th the comet passed its perihelion at a distance of about 750,000 miles from the sun's center, and within 300,000 of its surface, rushing through the coronal regions with a velocity exceeding 300 miles per second: it swept over 180° of its orbit in three hours and a half. Thus far we find in our lists of cometary orbits only four with so small a perihelion distance, viz., the comets of 1668, 1680, 1843, and 1880. (As to the comet of 1008 there is some doubt, because it was only observed for about three weeks, and its motion during that time was such that it answers almost equally well to either of two quite different orbits.) There are half a dozen others with perihelion distances between one and a half and five million miles, viz., comets of 1767, 1816, 1826, 1847, 1865, and 1870; and Wells's comet, which disappeared only a few weeks ago, is just outside that limit, with a perihelion distance of 5,675,000 miles. Now, as to the comets of the first class, we find that, excepting that of 1680, their orbits are extremely similar; their plane and direction of motion are almost exactly the same; the perihelion distances are nearly the same for all; and the axes of the orbits all point to the same part of space; they have all come toward the sun from the same region of the heavens, in the immediate neighborhood of the great star Sirius. In the little table below are given what are called the elements of their orbits: Ω, is the longitude of the node, i the inclination of the orbit to the ecliptic, π the longitude of the perihelion, and q the perihelion distance, expressed as a decimal fraction of the earth's distance from the sun; e is the eccentricity of the orbit; and the—in the last line denotes that the motion is retrograde. The orbits of the first two are
| 1668. | 1843. | 1880. | 1882. | |
| Ω | 357° 17’ | 361° 12’ | 356° 17’ | 345° 50’ |
| i | 35° 58’ | 35° 41’ | 36° 53’ | 38° 05’ |
| π | 277° 2’ | 278° 39’ | 278° 23’ | 276° 28’ |
| q | 0∙0047 | 0∙0055 | 0∙0059 | 0∙0076 |
| e | 1∙0 | 0∙99989 | 0∙99947 | 0∙99997 |
| Direction | — | — | — | — |
from the catalogue in Chambers's "Descriptive Astronomy"; that of 1880 is the orbit computed by Meyer, of Geneva, from the whole assemblage of observations, and that of 1882 is the last orbit computed by Mr. Chandler, of Cambridge, and may be found to need some correction when later observations come to hand. Fig. 4 shows in a rough way how these orbits lie in relation to the orbit of the earth, and how very long and narrow the comet's orbit is as compared with the circle described by the earth.
Now, the similarity between these orbits may be explained in two different ways. It might be accounted for by supposing that we have to do with different visits to the sun of a single comet, or that we have here a group or family of comets, very likely of common origin,
