round him in short periods, and at correspondingly small distances. The periods, approximately expressed, are 2, 3, 7 and 17 days, and the distances 6, 10, 15 and 21 times the radius of Jupiter.

The orbits of the satellites are found to coincide nearly, but not exactly, with the plane of Jupiter's equator. Hence, as both the plane of his equator and that of his orbit have but small inclinations to the plane of the ecliptic, the satellites can never deviate far from either of these planes. They, therefore, always appear to be, and to move forward and backward, nearly in a straight line which crosses the centre of the disc in the direction of the belts. The discovery of Jupiter's satellites by Galileo, was one of the first fruits of the invention of the telescope. They are visible with. a telescope of small power.

tance.

317. Disappearances of the satellites. Frequently, not more than two or three of the satellites are visible; sometimes not more than one; and one instance is stated to have occurred, when they were all invisible at the same time. The causes of these disappearances will be easily perceived by reference to Fig. 52, in which S represents the sun, J Jupiter, E the earth, EK a part of the earth's orbit, and abm the orbit of one of the satellites, in which it moves in the direction abm. Jupiter is so remote from the sun, and so large a body, that his shadow extends to an immense disA portion of it, only, is represented in the figure. Now, supposing the earth to be at E, and the satellite to be beyond Jupiter, in the part of the orbit from e to d, it must evidently be invisible, being hid by the planet. When it is between the earth and Jupiter, in the part of the orbit from e to f, its light becomes blended with that of the planet, and it is generally invisible; if, however, it happens to be directly between the observer and one of the belts, it may sometimes be seen with a powerful telescope, appearing as a bright spot on the belt. When the satellite enters the shadow at a, it necessarily becomes invisible, as it then ceases to receive light from the sun. This last phenomenon is called an eclipse of the satellite.

While the satellite is passing from m to n, it casts a shadow on the planet. With a telescope of high power, the shadow may be distinctly seen as it traverses the disc.

318. Eclipses of Jupiter's satellites. The eclipses of Jupiter's satellites are phenomena of very frequent occurrence. For, in consequence of the great size of the planet, the small distances of the satellites, and the small inclinations of their orbits to that of their primary, the three interior satellites suffer an eclipse every synodic revolution; and the fourth very rarely passes opposition without being eclipsed.

Both the beginning and end of an eclipse of the third or fourth satellite, or the immersion and emersion, at a and b, may frequently be observed from the earth; both taking place on the same side of the planet. This is, also, sometimes the case with the second. But the orbit of the first is so near to Jupiter, that its immersion and emersion can never both be seen; one or the other taking place behind the planet. This will be perceived by supposing an orbit to be described, much smaller than that in the figure.

It is evident, from inspection of the figure, that the eclipses take place to the west of the planet, while the earth is to the west of SJ, that is, before the opposition of Jupiter; and to the east, while the earth is in the other half of its orbit, or after opposition.

319. Revolutions and motions of the satellites. From the observed times of immersion and emersion of a satellite, the time it is in opposition to the sun becomes known; for this time must evidently be the mean of the two former. It, therefore, follows that repeated observations of the eclipses of a satellite serve to determine its mean synodic revolution. From this, the periodic or sidereal revolution is easily found.

From the mean sidereal revolution, the mean motion or angular velocity becomes known.

The orbits of the satellites differ but very little from circles, and, consequently, their true elliptical motions differ but little from their mean motions. The mutual actions of the satellites produce, however, some perturbations in their motions. These have been carefully investigated by Laplace and others; so that their true motions are now quite accurately known.

320. Curious relation in the mean motions of the first three satellites. If the mean angular velocity of the first satellite be added to twice that of the third, the sum will be equal to three

times that of the second. From this, it follows that, if from the sum of the mean longitude of the first, and twice that of the third, three times that of the second be subtracted, the remainder will always be the same quantity; and, from observation, it is found that this quantity is 180°. Hence, it also follows that the first three satellites can never all be eclipsed at once.

321. Use of the eclipses of Jupiter's satellites in determining the longitudes of places. As a satellite, on entering the shadow, loses its light, and on leaving regains it, the same immersion or emersion must occur at the same instant for different places, however distant from one another. If, then, the times of immersion or emersion, as reckoned at two different places, be accurately observed, the difference of these times must be the difference of longitude of the two places. Consequently, if the longitude of one of them is known, that of the other becomes also known.

The times of the eclipses, computed from tables which have been formed for the purpose,* are given in the Nautical Almanac, for the meridian of Greenwich. These computed times differ but little from the times observed at that meridian. If, then, an eclipse of one of the satellites be observed at any place, the difference between the observed time, and the time given in the Nautical Almanac, expresses the longitude of the place from Greenwich.

This very simple method of finding the longitude is not so accurate as some others. For, as the light of the satellite gradually diminishes, while it is entering the shadow, and gradually increases as it is leaving it, like that of the moon when entering and leaving the earth's shadow, the observed time of disappearance or reappearance of the satellite, must depend on the power and perfection of the telescope used, and, in some measure, on the eye of the observer.

322. Transmission of Light. The grand discovery that the transmission of light is not instantaneous, but that it requires time proportionate to the distance, is due to Roemer, a Danish astronomer, who deduced it from observations of the eclipses of Jupiter's satellites. In 1675, Roemer examined and compared observations

* The best tables of Jupiter's satellites are those computed by Damoiseau.

1

of the eclipses of the satellites which had been made during a number of preceding years. He found that the eclipses which happened about the time of Jupiter's opposition, when he was nearest to the earth, all occurred some minutes sooner than they should do, according to the averages of the intervals between consecutive eclipses of each satellite; and that, when Jupiter was near conjunction, and, consequently, most remote from the earth, they all occurred as much later than they should do, according to these averages. The deviations appearing thus to be connected with the planet's distance from the earth, it occurred to him, while seeking for their cause, that they could be explained by assuming light to be uniformly transmitted in time: that is, by assuming that, when any very distant phenomenon happens, a measurable interval of time, proportionate to the distance, elapses between the actual occurrence of the phenomenon and the perception of it by the observer. Pursuing the inquiry, he found that the deviations he had noticed would be completely accounted for, by allowing 8m. 13sec. for the transmission of light through the distance between the sun and earth. This, since the sun's distance from the earth is 95,000,000 miles, gives to light the amazing velocity of more than 192,000 miles per second.

This conclusion, with regard to the transmission of light and its great velocity, subsequently received complete confirmation by Dr. Bradley's discovery of the abberration of light (131).

323. Rotation of Jupiter's satellites. From very frequently repeated observations of Jupiter's satellites, it has been ascertained that they are subject to marked periodical fluctuations with regard to brightness; and that the periods correspond respectively with the periodic revolutions of the satellites. Hence, it has been inferred that each satellite, like our moon, revolves on its axis in the same time that it revolves round the planet.

SATURN AND HIS SATELLITES AND RINGS.

324. General remarks. Saturn is a large planet, being next in size to Jupiter, and not greatly inferior. He is so remote from the sun, in comparison with the earth, that his apparent diameter is not subject to much variation. Its mean value is about 17". Consequently, Saturn, though so remote, is, from his great size, a tolerably conspicuous object. He shines with rather a pale white light.

In addition to his eight satellites (9), Saturn is distinguished from all the other planets by being surrounded, at some distance, by two broad, flat, circular rings, situated in the same plane, and concentric with the planet and with each other.

325. Saturn's period, distance, &c. Saturn revolves round the sun in about 29 years at the distance of 905 millions of miles. His diameter is about 79,000 miles, and his bulk nearly 1000 times that of the earth. He revolves in 10h. 29m., about an axis making an angle of 28° 40′ with the axis of the ecliptic.

326. Saturn's Rings. The rings of Saturn are opaque bodies, shining like the planet, by reflecting the light of the sun. This follows from the fact that they are observed to cast a shadow on the side of the planet next the sun, and to be shaded by it on the opposite side. Fig. 53, represents Saturn surrounded by these. singular appendages; the body of the planet being striped by dark belts somewhat similar to those of Jupiter, but broader and less strongly marked.

From micrometrical measurements, it has been ascertained that the distance from the surface of Saturn to the inside of the nearest ring is a little over 19,000 miles; the breadth of this ring is about 17,000 miles; the interval between the two rings is 1,800 miles; and the breadth of the exterior ring is about 10,600 miles. The entire diameter of the exterior ring is 176,000 miles. The rings are extremely thin; their thickness, according to Sir J. Herschel, does not exceed 100 miles.

When the rings are examined with telescopes of moderate power, they appear as one, the interval between them not being percep