132 



Dr. S. W. Richardson on the Magnetic 



fig. 6. The box containing the ring was placed in a vessel of 

 water which was heated from below by a bunsen-burner. 

 The water was automatically kept in movement by means of a 

 stirrer driven by a current "of air passing through an asbestos 

 cylinder and impinging upon a wind-wheel Y. 

 1 To obtain temperatures between that of the atmosphere and 

 that of ice, the source of heat was removed and small pieces 

 of ice were added to the water from time to time until the 

 desired temperature was obtained. 



V = wind-wheel. 

 T = stirrer. 

 B = copper "box. 

 G = glass tube. 



L = leads. 



S = source of heat. 



A = asbestos cylinder. 



(3) Experiments at Temperature.? beloic Zero. 



The same box and attached tube was used as in the last 

 series of experiments. It was found that, at the low tempe- 

 ratures used, ice formed to the thickness of a centimetre 

 or more along the metal leads, owing to the condensation 

 of the moisture in the air. and thus the insulation was 

 endangered. 



To overcome this difficulty the following arrangement was 

 adopted (fig. 7). The secondary and compensating leads were 

 surrounded by an indiarubber tube T, the free end of which 

 was carefully closed. Some portion of this tube passed down 

 into the glass tube G. This was then filled with melted paraffin. 

 By this means the access of moisture to the leads was confined 

 to that in the indiarubber tube, and the formation of ice upon 

 them was almost entirely stopped. The closing of the glass 



