Magnetic Permeability of various Alloys of Iron. 101 



Kelvin's graded galvanometers.* One great advantage of this instrument was 

 the ease with which its range could be altered by sliding it to or from the solenoid 

 on a grooved board, or by the addition of an over-head magnet, by means of which 

 the strength of the field at the magnet could be increased by a known amount. 

 The field-magnets usually employed on the instrument were too strong for our 

 purpose, and we had two smaller ones specially made. The strength of these 

 were found to be 0"875 and 2 -927 C. Gr. S. units respectively. These, with the 

 Earth's field, as carefully determined at the place of observation, viz. 0'175 

 C.G. S., gave us three convenient grades. The standard distance of the magneto- 

 meter needle from the upper pole of the rod under test was 45 cms. This distance, 

 in the case of a few of the less permeable steels, was reduced to either 20"9 cms. 

 or 14*75 cms. ; these distances being chosen because the deflections, when divided 

 by 5 or 10, respectively, corresponded to those at the standard distance. This 

 ratio, obtained by calculation, was verified by actual experiment at the different 

 distances. 



The test-rod was sujoported verticall}', as shown in fig. 7, inside the 

 magnetising solenoid C. The latter was 120 cms. long and 1*7 cms. mean 

 diameter, and wound with 2886 turns of No. 18 B.W. G. double-cotton-covered 

 copper wire, in four layers. This was sufficient to carry a current of 10 amperes 

 for a short time without undue heating. The resistance of the coil was 2' 15 

 ohms, and the magnetising force was 30'2 C.Gr.S. units per ampere. 



In order to neutralize the effect on the magnetometer of the magnetic 

 field due to the solenoid, a compensating coil, C',t was introduced into the 

 circuit, so that the main current passed through it as well as the solenoid. 

 By trial, on moving the compensating coil to and fro, a point of neutralization 

 was found, so that, when no test-rod was present, the magnetometer needle 

 remained unaffected whatever strength of current traversed the solenoid. Any 

 magnetisation of the rods due to the vertical component of the Earth's magnetic 

 field was neutralized by a single layer of wire round the solenoid, through which 

 a small independent current from a single cell was sent by the circuit W. The 

 strength of this current was adjusted to suit the particular rod under test by 

 means of the rheostat R', or it could be intercepted altogether by the plug key K', 

 as it was not required in the specimens of low permeability. 



The main magnetising current was obtained from 1 to 5 cells of a storage 

 battery B, and after traversing the plug key K, and circuit W, entered a series of 



*■ This magnetic system is composed of four small magnetic needles, about 1 cm. long each, with their 

 poles turned in similar directions, and supported on an iridium-tipped point and sapphire cap. The 

 pointer is an aluminium index, about 9 cms. long, working over a tangent scale. 



f In fig. 7 this is also marked C by mistake. The relative distances of the compensating coil and the 

 magnetometer from the solenoid were much greater than would appear from the diagram. 



Q2 



