XX11 



as occurs in iron as it passes from the non-magnetisable to the mag- 

 netisable state. 



The behaviour of iron near this critical state was the subject of 

 Hopkinson's next magnetic paper.* It was shown that for small 

 magnetising forces the permeability of iron increases very rapidly 

 as the critical temperature is approached, and then very suddenly 

 disappears, also that the critical temperature is marked by a sudden 

 change in the coefficient of resistance, and that it is the point at 

 which recalescence occurs. The evolution of heat in recalescence was 

 measured, 



The same methods of inquiry were applied in the following year to 

 certain alloys of nickel and iron, which were found to be capable of 

 existing, throughout a wide range of temperature, in two states, one 

 magnetisable and the other not, The state changed from non- 

 magnetisable to magnetisable when the alloy was cooled somewhat 

 below C., but did not change back again until it was heated to 

 nearly 600 C. In the non-magnetisable state the nickel steel was soft 

 and ductile ; in the other state it was hard. Equally marked differ- 

 ences were found in respect of electrical resistance. A later series of 

 experiments deal with time-lag in the process of magnetisation, 

 especially on the influence which the electric currents induced in the 

 iron by magnetisation have in retarding the acquirement of magnetism.! 

 The growth of magnetism was observed in a very massive iron core, 

 by means of exploring coils buried at various places in its substance, 

 and the results were applied to determine the appropriate thickness of 

 laminated iron in transformers subjected to periodic reversals of 

 magnetism. These experiments formed the subject of a Friday 

 evening discourse at the Royal Institution, which concluded with a 

 remarkable speculation as to the possibility of terrestrial magnetism 

 being due to currents in the material of the earth sustained by its 

 changing induction but gradually dying away. 



In these and others of his later researches Hopkinson worked in 

 co-operation with Mr. E. Wilson, his assistant at King's College, 

 and the results were published in their joint names. The authorities 

 of King's had invited Hopkinson, in 1890, to assume the direction of 

 the Siemens Laboratory at King's with the title of Professor of Elec- 

 trical Engineering. The post made no considerable demand on him as 

 a teacher, but it gave him the use of a laboratory and the opportunity 

 of suggesting to students subjects of research. He was, moreover, able 

 to place a number of the King's College students in engineering- 

 situations, and the uniform success of the young men he favoured in 

 this way showed that he exercised his patronage with rare judgment, 



* Phil. Trans.,' A, 1889. 



f 'Phil. Trans.,' A, 1804; ' Inst. Elect. Eng.,' 1895. 



