154 
SECOND REPORT - 1832 . 
on Laplace’s theory, were published by Bouvard in 1808, but 
were soon suppressed, as it was found that in Burckhardt’s addi¬ 
tion to Laplace’s theory, several terms had been applied with the 
wrong sign (in consequence of mistaking the perihelion for the 
aphelion). A new set of Tables was therefore published in 1821, 
with the improved theory, and founded on ah the good obser¬ 
vations of modern astronomy. In discussing these, values are 
obtained for the masses of Jupiter, Saturn, and Uranus. 
In the Ast. Nadir. No. 97 and 139, are micrometrical mea¬ 
sures of Jupiter and Saturn by Struve. He determines the 
flattening of Jupiter to be yUy, and the inclination of Saturn’s 
ring to the ecliptic to be 28° 52 Bessel ( Berlin JEphemeris 
1814 and 1829,) had made it about 28° 22'. These values are 
considerably less than that formerly received (about 81° 20'). 
In No. 189 Ast. Nadir, are measures of Saturn by Bessel, with 
a divided object-glass. 
In the Phil. Trans. 1805, 1806, and 1808, Sir W. Herschel 
gave observations of Saturn’s figure. It appeared that about 
latitude 45° the planet projected above the elliptic form. I 
think it worth mention that I have myself witnessed an instance 
in which a person, who had never heard of this observation, 
on seeing the planet very distinctly, made spontaneously the 
same remark. I have many times seen the planet with extreme 
distinctness, and have on one occasion thought that it certainly 
had this shape ; and on another, have been equally convinced 
that it is rather flattened at latitude 45°. The shape assigned 
by Sir W. Herschel (S ee Monat. Corr. vol. 15, and Cambridge 
Transactions , vol. 2) cannot be reconciled with theory. 
In 1821, Bouvard published Tables of Uranus (in the same 
volume with those of Jupiter and Saturn). With respect to 
this planet a singular difficulty occurs. Seventeen observations 
of Uranus have been found in the observations of Bradley, 
Mayer, &c. (for discussions of which see the Zeitsdirift, the 
Conn, des Temps , &c.) ; and since its discovery as a planet in 
1781, observations have not been wanting in any year. Now 
it appears impossible to unite all these observations in one ellip¬ 
tic orbit, and Bouvard, to avoid attributing errors of importance 
to the modern observations, has rejected the ancient ones en¬ 
tirely. But even thus the planet’s path cannot be represented 
truly ; for these Tables, made only eleven years ago, are now in 
error nearly half a minute of space. 
Delambre’s new Tables of Jupiter’s satellites (for eclipses), 
published in 1817, were founded on all the observations that he 
could collect from 1662 to 1802, and on Laplace’s theory; and 
will probably want little alteration for some years. It is to be 
