228 MR. S. BIDWELL ON CHANGES PRODUCED BY MAGNETISATION IN 



employed) in order to arrive at a more definite determination of those portions of the 

 curves where greater detail seemed to be required. The corresponding values of 

 magnetising force and change of length thus obtained are given in Table VII., and 

 indicated by dots in the diagram fig. 6. 



TABLE VII. 



Fis*. 6. 



Referring to this diagram, it will be seen that in the cases of iron and cobalt the 

 coincidence of the two sets of observations is as close as could be expected. The 

 earlier portions of these curves have, in fact, been drawn through the dots without 

 any regard to the position of the crosses. But for nickel the two series of results 

 diverge very considerably. The curve is here drawn as smoothly as possible through 



