these occasions have been from 300 to 800 times; sometimes one, and sometimes another being used, according to the states of the air or moon, or other circumstances.
The observations here recorded, were made principally from the year 1787 to 1798 inclusive, and they are given in the order in which they occurred with all the phenomena noticed each time, and notes of correction added from subsequent observations respecting stars mistaken for satellites, or satellites wrongly numbered. For Dr. Herschel always added a configuration to his descriptions, in order to avoid mistakes, and in general made, by previous estimate, a sketch of the places where known satellites might be expected; some misconceptions were unavoidable, in consequence of the interruptions to his observations from cloudy or moon-light nights, as well as from small stars that unexpectedly interfered.
From all the observations thus made, and from accurate measures taken by different micrometers, the author infers the nodes of the two first satellites to have nearly the same longitude of 16512°; that their inclination is also the same, 78° 58'; and the motion of both, from their ascending node to the greatest elongation, retrograde. The periodic time of the first is found to be 8d 16h 56m 5s; and that of the second 13d 11h 8m 19s; their distances at greatest elongation 36" and 48" respectively.
To these determinations respecting the orbits of the satellites, Dr. Herschel adds his estimate of the distances to which they must move from the body of the planet before they become visible by his 20-feet telescope. The first requires to be at more than half its greatest elongation. The second also becomes invisible when it is within half its greatest elongation. If there be an interior satellite, as large as the first, he imagines it would be visible through so small a part of its orbit, as not to be seen for two nights together.
Notwithstanding this difficulty, the author thinks he has seen an interior satellite. And with regard to exterior satellites, though nothing has been correctly ascertained, the number and positions of different objects recorded is such, that the author enumerates as far as a supposed sixth satellite.
The battery with which these experiments were made, has 21 cells, each containing about 40 gallons of water, to which was added a mixture of nitric and sulphuric acids, at various intervals, beginning with 140th, and ending with 120th part of the water. Into each cell was immersed a zinc plate, with a pair of copper plates, one on each side, but connected together at the bottom, and also duly connected to the zinc in the cell adjacent. Each plate had 16 square feet of surface.
The first series of experiments were on the comparative liability of
