4G8 



PROFESSOR KNOTT ON THE STRAINS PRODUCED IN IRON, 



phenomenon has a strong resemblance to hysteresis; but, as proved by the other 

 experiment, in which the return value is the greater, it is merely a resemblance and 

 nothing more. That the phenomenon does not depend only on the iron, but has 

 something to do with the brass tube, is proved by the fact that there is not the slightest 

 evidence of its existence in the experiment for measuring the change of volume of bore. 

 Thus, compare with the foregoing tables the following : — 



A III., November 12th, 1897. 

 Change of Volume of Bore. 



Field. 



Field on. 



Field off. 



531 

 403 

 302 



100 



48 



+ 47 

 + 47 

 + 46-5 



+ 47 



+ 14 



-47-5 

 -47-8 

 -47 



-47 



-13 



In this case the capillary was in connection with the interior of the tube, and 

 a positive reading means a contraction of the volume. 



The complete absence of the peculiar effect in this last case quite disposes of 

 any attempt at explanation in terms of hysteresis or possible change of temperature. 

 The effect is as if at every make and break of the current in the magnetizing coil 

 the brass tube containing the iron were permanently increased in internal capacity, or 

 as if a certain quantity of water in the tube were removed from it. It is conceivable 

 that a sudden increase of volume of the inclosed iron might push out a small quantity 

 of water so that a positive reading might be followed by a greater negative reading, 

 as in the experiment of November 1 5th ; but it is altogether inconceivable that a 

 contraction of the iron should be accompanied by a like pushing out of water, as 

 would have to be the case were the explanation to hold good for the first experiment of 

 November 12th. 



One early suggestion was that the slight difference of diameter between the iron 

 tubes and the bore of the brass tube might produce a certain constraint on the film of 

 water between. But it was found on trial that the peculiar effect persisted in the case 

 of an iron bar whose diameter was made distinctly smaller than the diameter of bore of 

 the brass tube. 



Thus, by a process of exclusion, we seem to be driven to the view that the brass tube 

 must experience, when the field is established, an abrupt change of volume from which 

 it does not immediately recover, or from which it recovers slowly when the field is 

 removed. Care was taken to have the iron tube as nearly as possible centrally placed 

 in the magnetizing coil. But no doubt there was some lack of perfect symmetry, so that, 





