On the Magnetic Deformation of Nickel. 45 



In offering the following results, I wish again to say that the 

 measurements were made in the Physical Laboratory of the Univer- 

 sity College of North Wales, and that I am greatly indebted to 

 Professor A. Gray for providing me with the apparatus, and allow- 

 ing me the time necessary for the experiments, and for many valuable 

 suggestions. 



Preliminary experiments were first made to find the nature of the 

 influence of temperature on the magnetic contraction. The appa- 

 ratus was essentially the same as described in the former paper, 

 with the addition of a spiral tube and burner for heating the water 

 just before it entered the coil. The temperature of the water on 

 entering and leaving the coil was indicated by two thermometers 

 inserted in capsules through which the water flowed. When the 

 water was warmed and allowed to flow for some time, the two 

 thermometers indicated steady temperatures differing generally by 

 about 2 C. 



The change of length was magnified by the same lever arrange- 

 ment as before, and observed by a telescope, scale, and mirror. A 

 few trials showed that it would be impossible to make any measure- 

 ments when the water was warmed unless precautions were taken to 

 remove air dissolved in the water, for this, on being liberated in the 

 heater, ascended in bubbles through the coil and disturbed the appa- 

 ratus so much that no readings could be taken. To prevent this, the 

 water was made to flow into a vessel through a pipe near the top, 

 and escape by a pipe near the bottom, whence it proceeded to the 

 spiral heater. A tall glass tube projected upwards from the vessel, 

 and the vessel was heated by a Bunsen flame. The dissolved gases 

 were thus liberated in this vessel, and, collecting at the top, escaped 

 at intervals up the glass tube. By this means the apparatus was 

 rendered steady enough to admit of readings at moderately high 

 temperatures. Further, by regulating the supply the water could be 

 made to stand at any level in the glass tube. Thus the flow of 

 water through the coil, and hence the temperature of the coil could 

 be controlled. 



The glass tube was about 6 feet long, and by raising the level of 

 water in it through 1 foot the temperature of the coil was lowered 

 by about 1 C. 



Observations of the change of length were always made at night, 

 generally between 10 P.M. and 2 A.M., the apparatus never being 

 sufficiently steady in the daytime. 



The nickel wire was the same as that used in the former experi- 

 ments, of length 83'4 cm. between the terminal brass pieces, and 

 mean diameter 1*65 mm. This was re-annealed before the present 

 measurements. 



A number of observations were then made of the change of length 



