is at once satisfactorily explained.  When the satellite appeared to pass on to the disc, and to be lost in the light of the planet, it would for some time, proportional to the depth of Jupiter’s atmosphere, have behind it a background of clouds only, it would not have entered upon the actual disc of the planet.  If then these clouds were suddenly absorbed, the atmosphere behind the satellite would become transparent and invisible, the background would be gone, and the satellite would reappear.  In the case of the occultation witnessed by Messrs. Gorton and Wray, the satellite would at first be hidden by cloud only, and would reappear if the cloud were removed.  Such seems to be the true explanation of these hitherto mysterious phenomena.  That they could not have resulted from any alteration in the motions of the planet or the satellite is evident.  Such an alteration would have been instantly detected, since the places of both the planet and the satellites are computed years in advance, and any such change would at once have thrown out all these computations.

Assuming that this is the true solution of the mystery, we are enabled to form an approximate estimate of the extent of the atmosphere of Jupiter.  The time between the first and second disappearances does not seem to have been accurately noted.  Admiral Smyth’s account makes it 16 or 17 minutes; but if we estimate it at 15 minutes only, and if we further assume that the second disappearance was upon the actual disc of Jupiter, and not upon a lower stratum of clouds, we shall be safe from any risk of exaggeration.  The probability seems to be that the second disappearance was caused not by the disc, but by the formation of a fresh body of cloud, as it was not gradual, as in the first instance, but sudden.  We shall then only have an estimate which cannot be greater, but may be much less, than the true value.

The mean distance of the second satellite from the centre of Jupiter is in round numbers 425,000 miles, and consequently the circumference of its orbit is 2,671,000 miles.  The satellite travels through this orbit in about 86 hours, which gives a horary velocity of 31,400 miles, or 7850 miles in 15 minutes.  This then is the least possible depth of the atmosphere of Jupiter. [Footnote:  For the direction of the motion of the satellite would be at right angles to the line of sight.] The whole diameter of Jupiter, atmosphere and all, is 85,390 miles.  Deduct from this 15,700 miles for the atmosphere, and we have for the diameter of the solid nucleus rather less than 70,000 miles.  The height of the atmosphere is therefore not less than three-fourteenths of the radius of the planet, and may be much greater.  The extent of the atmosphere, combined with the rapidity of rotation, accounts satisfactorily for the great apparent polar compression of the planet.  Another inference is that the density of the planet must exceed the ordinary estimate in the proportion of two to one.