432 Mr. E. H. Hall on the Rotation of the — 
stant during these experiments. It is possible, however, that 
it diminished one or two per cent. during the series. The 
error introduced in this way is of slight consequence for the 
present purpose. Again, the direct effect of the electro- 
magnet in disturbing the needle of the Thomson galvanometer 
was neglected, together with certain other particulars which 
may have introduced a constant error of several per cent. into 
the values obtained for the rotatory power. 
| 
No. of Exp.| Date. | Temp. | M. C. Rae 
1.  |Dec. 22,1883.) 3° 3463 | 1251 | +2092x10-% 
2. F 13 i 1243 | +2061 
a ‘ 2076 
2, i 18 | 10887 | 2 oa 
4. ‘ 18 ; 1244 | 2390 
18 2415 
| 
Decrease of R. P. for fall of 1° C. is approximately % per cent. 
Bismuth. 
I can only confirm what Prof. Righi* has already published 
concerning the rotational effect in this metal. The rotational 
power appeared to be about 260 times as great numerically as 
that of tempered steel, and of the opposite sign. 
The slice of bismuth used was obtained, shaped, and mounted 
in much the same manner as the slice of cobalt already de- 
scribed. ‘The length and breadth of the bismuth cross were 
somewhat greater than the corresponding dimensions of the 
cobalt and the thickness was about 1 millim. 
It seems probable that a thin slice of bismuth, properly 
prepared and mounted, will come to be a valuable instrument 
for measuring the intensity of strong magnetic fields. 
Antimony. 
The powerful effect observed in bismuth suggested an ex- 
amination of antimony. A slice of this metal, much like the 
slice of bismuth just mentioned, was prepared. The cross 
suffered in shape a good deal in making the adjustment. The 
final average thickness in the central portion was estimated 
at 1:2 millim. 
* Acc. dei Lincer Transunti, June 1883. 
