period could not differ much from 2440 days, and that an ellipse with a major axis corresponding to this time, would represent equally well the observations of 1805 and 1826.

The comet of 1826 was observed by Dr. Olbers and M. Gambart until the end of April. It is generally described as a small circular nebulosity, destitute of tail or nucleus, and a very little brighter towards the centre than at the edges. Soon after the disappearance of the comet, Professor Santini, of Padua, undertook the determination of the effects of planetary disturbances on the ensuing return. He ascertained that the most probable value of the periodic time in 1826 was 2455.176 days and that the comet had arrived at its perihelion point on March 18th, at ten P.M., Paris time. He then calculated the amount of perturbation due to the attraction of the Earth, Jupiter, and Saturn, and found that the next revolution would be shortened 10.023 days by their combined influence, so that the comet should again arrive at perihelion about two o'clock on the morning of the 27th of November, 1832. The Baron Damoiseau also entered upon a similar investigation, and was led to infer that the return of the comet would be accelerated by 9.664 days, a result not very widely differing from that of Professor Santini.

In a paper by Dr. Olbers, published at the beginning of the year 1828, the attention of astronomers was directed to the very close approach of the orbits of the earth and comet at the descending node of the latter. In 1805 and 1826 the radius-vector of the comet at the nodal passage was greater than the

E

earth's, and consequently it passed outside our orbit. But in 1832 Dr. Olbers found that the comet at this moment would be nearer the sun than the earth, and would therefore pass inside our annual path, but at a distance from it of only 0·00033, or less than five terrestrial semi-diameters. Now, on the 8th of December, 1805, when the comet of Biela was near the earth, Dr. Olbers found the apparent diameter subtended an angle of 40', whence he concluded that the real semi-diameter of the nebulosity was at least 5.25 radii of the earth, and it thus appeared certain that if the extent of the cometic atmosphere in 1832 were as great as in 1805, and if Damoiseau's calculations were rigorously accurate, a portion of the earth's orbit would be within the nebulosity of the comet at the nodal passage in 1832. The orbital arc, or arc of true anomaly between the descending node and perihelion amounted to 41° 45', which the comet would require 29.0 days to traverse, and as the calculations had fixed the arrival at perihelion about midnight on the 27th of November, it was thus inferred that the passage through the descending node would take place on the evening of the 29th of October. The heliocentric longitude of the point of the comet's orbit which lies nearest that of the earth is 68° 10′, and the earth could not reach this point until the morning of the 30th of November, or one month after the comet's passage by it. These results, of course, showed that there was no ground for alarm, at least in 1832, as the two bodies would pass through the dangerous neighbourhood so long after each other. If the perihelion passage had occurred at eight P.M. on the 28th

of December, an extremely near approach, if not a collision, of the earth and comet must have taken place on the last day of November.

Though astronomers were satisfied the comet would not approach within many millions of miles from the earth, it was not so with the general public. Considerable alarm was excited on the continent as soon as Olbers' results were known, and the comet of 1832 was even anticipated as the destined agent in the destruction of our globe. It was argued that if a retardation of about one month in the arrival at perihelion should take place, the most disastrous consequences must follow-a line of reasoning which, though possibly correct in itself, was altogether inadmissible under the circumstances. The periodic time corresponding to the ellipse actually described by the comet at its visit in 1826 was known, without a greater error than one day, and the effect of planetary attraction had also been determined with all requisite accuracy. The date fixed for the perihelion passage, November 27th, could not therefore be well subject to a larger error than one day at the utmost. All fear of collision or dangerous proximity was evidently groundless.

The first glimpse of the comet at its reappearance was obtained by the observers of the Collegio Romano at Rome, on the 23rd of August. It was observed by Sir John Herschel, with his twenty-feet reflector, at Slough, on the 23rd of September, but at this time neither Professor Harding, with a 10-feet reflecting telescope, nor M. Nicolai with a 41-feet achromatic, by the celebrated Fraunhofer, could dis

tinguish it. The excessive faintness of the comet may be readily conceived. About the end of the third week in October it became more generally visible, and was last seen at the Cape of Good Hope, on the 3rd of January, 1833. It was always faint, with but little central condensation, and was at no time perceptible without a good telescope. The comet arrived at perihelion only twelve hours before the computed time,—a much closer fulfilment of prediction than could have been expected. Professor Santini, on a revision of his calculations, and after allowing for the attraction of Venus and Mars, which he had previously neglected, ascertained that the time elapsed between the perihelion passages of 1826 and 1832, was 2444-7027 days, his earlier computations having given 2445-1528 days. He then investigated the effects of planetary disturbances on the ensuing appearance, and found it would retard the comet 1d 17h, so that the perihelion passage would occur on the 23rd of July, 1839, about noon, Greenwich mean time. The longitude of the comet in perihelion, as seen from the sun, is 110°, which differs only 10° from the geocentric longitude of the sun on the 23rd of July; and the comet's motion being direct, it is always too near this luminary to be easily detected when the perihelion takes place about this time. It passed through these parts of space in 1839 without being perceived in any part of the earth, and the next return, in 1846, was therefore anticipated with the greater interest. The calculations of Professor Santini, to whom we are indebted for nearly the whole of our knowledge of the comet's movements since 1826, indicated an acce

leration of the ensuing arrival at perihelion by 31.884 days, the influence of the planet Jupiter being very considerable, not only on the periodic time, but likewise on the inclination of the plane of the orbit to that of the ecliptic, which was diminished 37′ in the interval between 1839 and 1846. The perihelion passage was fixed for about nine p.m. Greenwich time, on the 11th of February, 1846, and it was found the comet would remain visible a long time, and thus afford an opportunity of correcting the theory of its motion, by long-continued observation.

The powerful telescopes which are now happily found in many of the European observatories were employed in the search for the wanderer, as the time of its return drew near. It was discovered the same evening, November 28th, by Professor Encke at Berlin, and Signor De Vico at Rome; Professor Challis saw it on the 1st of December with the great Northumberland telescope at Cambridge, but it was not generally recognised till the third week in December, and was last seen on the 27th of April at the observatory of Bonn.

A phenomenon which created no little astonishment amongst astronomers, took place at this reappearance of Biela's comet. When it first became visible in the large refractors of Berlin, Cambridge, &c., it presented a faint nebulosity, almost if not perfectly circular in form, with but slight condensation towards the centre. On the 19th of December it was remarked to be somewhat elongated or pear-shaped, but this circumstance merely induced a passing notice, such distortions being not unfrequently noticed in telescopic