repetition of M, Arago's experiments on magnetism^ &c. 483 
1 rev. 2 rev. 3 rev. 4 rev. 5 rev. 
28.2 39.7 48.4 56.3 63.0 
which require the exertion of an accelerating force more than 
double of that developed in the last trial. 
The bismuth was now melted out, and the cuts being 
carefully washed with melted tin, were filled with fresh tin, 
which was allowed to fix, and the disc being trimmed, and 
replaced, the times were now found to be 
21.7 30.8 38.0 43.5 48.7 
The restoration of energy, as in the case of the copper 
disc is here very manifest, the times of rotation being nearly 
reduced to their original magnitude. The comparison of 
these, reduced by the formula/ = 1000000 -1 , and the means 
of the five results taken in each case, gives for the accele- 
rating forces 
Brass, uncut _ - - 1.00 
cut - - - 0.24 
soldered with bismuth 0.53 
soldered with tin - 0.88 
The effects of soldering with lead and with fusible metal 
were also tried, and found to be both represented on the same 
scale by the same fraction, viz. 0.85, being but very little 
inferior to tin. 
22. When the soldering is imperfect, the effect in restoring 
the magnetic action is proportionally weaker, but the influence 
of ever so small a free metallic communication is sensible. 
23. A disc of lead cut in 8 radii as above was found to 
make one revolution in 58 ‘.3. It was then wetted so as 
to fill the cuts with sulphuric acid, and the time of revolution 
was found to be 57.3 ; so that the influence of sulphuric acid, 
3 R 
Copper uncut - 1.00 
cut - - 0.20 
soldered with tin 0.91 
MDCCCXXV. 
