magnetical properties in steel and iron by percussion. 24,9 
was increased ; thus a quarter-inch cylindrical bar of steel 
five inches long, after receiving 20 smart blows, produced 
a deflection of the needle, at the distance of three inches, of 
13 0 , and lifted 6j grains. Another piece of the same bar 
7-| inches long, after similar treatment, produced a deviation 
of 2 4 0 , and lifted 45 grains ; and a third bar of the same 
kind 12 inches long, after twenty similar blows, occasioned 
a deviation of the compass of 33 0 , and easily lifted 88 grains. 
The shortest bar, it was observed, received the full effect by 
the two first blows ; but the others continued to increase in 
energy as the percussion was continued. These bars did not 
receive a power equal to that first used ; the cause was pro- 
bably their greater hardness. 
3. A strong magnet properly tempered, was injured in 
whatever position it was hammered ; but most rapidly when 
the north pole was upward. After no farther diminution of 
its magnetism could be produced with the south end upward, 
a quick loss of power was effected by hammering it with the 
north pole upward (3 6). But after the magnetism had been 
reduced to a certain extent, by hammering in both positions, 
the power became nearly stationary ; so that on striking it in 
any position with the same hammer, very little change of 
intensity occurred (3 c. ). 
Besides these results, I may mention the effect of percussion 
on soft steel magnets ; on soft iron not magnetised ; and on 
cast iron. 
4. A soft steel magnet capable of lifting upwards of loco 
grains, when placed vertically upon the poker with its north 
pole upward, had its magnetism destroyed by five blows. 
5. A bar of soft iron, of the same size and form as the 
MDCCCXXII. Kk 
