10 Barretr, Brown, & Haprintp—On the Magnetic and Klectric 
being similar to that of a curtainring. ‘The densities at 12° C. were also taken as 
shown in the last column. 
The rings were each wound with a primary coil of 598 turns in three layers, 
and a secondary coil of 200 turns on the S.C.I. ring, and 190 turns on the 898 E 
ring. A Crompton bifilar ballistic galvanometer was employed, and the value of 
its deflections was determined by an Earth inductor as well as by a standardising 
coil. The Earth inductor was wound with 442 turns of insulated wire, the mean 
radius of its coils being 13 cms., the horizontal component of the Earth’s field in 
which the inductor was placed was found to be 0:1714 in C.G.S. units; this value 
was checked several times. The standardising coil, which was used to check the 
calibration of the galvanometer by the Earth inductor, consisted of a uniformly 
wound solenoid 61:5 ems. long, having 2275 turns of wire on the outside, and in 
the middle of this primary coil a secondary coil of 200 turns of silk-covered wire 
was wound. The value of B per scale division thus found was slightly different 
for each ring, owing to their slightly different dimensions, &c.; each scale 
division of the galvanometer corresponded to an induction B, of 13°21 lines per 
sq. cm. for the 8. C. I. ring, and 11°95 for the 898 EK ring. 
The result of a long series of experiments with these rings is given in the 
cyclic curves shown in Plate III. It will be seen that these curves closely resemble 
those in Plate II., obtained from the slender rods by the magnetometric rod. The 
specimen of silicon-iron was not quite the same in the two cases. The rod and 
rings were made from different castings which, upon analysis, were found to have 
the percentage composition given in the next table; the specimen of Swedish 
charcoal-iron, S.C. I., had exactly the same composition in both cases, and is 
here given for the sake of comparison. ‘The iron is found by difference ; traces 
of manganese, phosphorus, &c., were also present. 
TABLE V. 
Si. C. Fe. 
Sy Oe Jog 5 . 0:07 0:03 99:9 
898 E rod, . : 2°50 0:20 97°3 
» ring, : 2°87 0°20 96°9 
These experiments afford an opportunity of testing the trustworthiness of the 
magnetometric method after applying the correction requisite for a rod of 255 
diameters. 
