4 Dr. J. G. Gray on the Magnetic Properties of a 



carbon with great readiness, special precautions had to be 

 taken to prevent admission of this element. The alloying 

 was conducted in a " Salamander " crucible under a layer of 

 fused barium chloride ; charcoal cannot be used if the per- 

 centage of manganese present is appreciable. The metal was 

 cast in dry sand moulds lined with whiting. 



The specimens were tested at room-temperature and at 

 — 190° C. (when immersed in boiling liquid air) in each of 

 the following conditions : (1) as cast, (2) after annealing at 

 900° C, (3) after quenching at 900° C. Condition (2) of 

 the metal was attained by imbedding the specimen in kaolin 

 clay contained within an iron tube, and enclosing the whole 

 in a Fletcher gas furnace. In this way the specimen could 

 be brought to 900° C, and cooled very gradually whilst out 

 of contact with the atmosphere. After testing the specimen 

 in the annealed condition it was brought to a temperature 

 of 900° C. and plunged vertically into ice-cold water ; it was 

 then in condition (3). 



The magnetic tests were made by means of a magnetometer 

 specially designed by Dr. A. D. Ross* and the writer for 

 the carrying out of investigations of this kind. The main 

 features of the instrument (fig. 1) are the simple and novel 



Fi*. 1. 





—ft— 



B 



T 











o 





o 







r~ 



— 1 













L 







M 

 5==?. 



c 3 



c, 





H 



c 8 



Plan of the magnetometer. 



methods introduced for rapidly securing accurate compen- 

 sation for the effect at the needle produced by the field of 

 the magnetizing solenoid. The magnetometer proper M, the 

 solenoid H, the compensating coils l9 C 2 , 3 , and the lamp 

 L for showing- the movement of the magnetometer needle, 

 are all mounted on a strong mahogany base-hoard, and move 

 freely in channelled beds. They can be firmly fixed by means 

 of friction clamps; (\ is the principal compensating coil, C a 

 a coil for fine adjustment, while a removes any error due to 

 wan! of proper alignment of (\, C 2 , and the solenoid. The 

 method of adjusting the apparatus for use is as follows. 

 * Phil. Mag, July 1909. 



