20 Mr. H. Tomlinson on the Sudden Loss of the 



at temperatures ranging from 750° 0. to 770° C, the higher 

 the greater the magnetizing force. The temperature at 

 which the iron begins to suddenly lose its temporary magnetic 

 susceptibility is about 680° 0. 



The author*, in a paper on the recalescence of iron, 

 was led by his experiments to express a strong conviction 

 that with iron there are two critical temperatures at which 

 profound changes take place in the molecular architecture 

 of the metal and at which heat becomes latent. The lower 

 of these critical temperatures he fixes at 550° C. and the 

 higher at 1000° C. Previously to this, but unknown to the 

 author, Pionchonf, by most carefully conducted calorime- 

 tric researches, had also found in iron two critical tempera- 

 tures at which heat becomes latent, namely, 660°C.-720°C. 

 and 1000° C.-1050 C. He fixes the value of the latent heat 

 at the lower temperature at 5*3 (about j^th of the latent heat 

 of ice), and that at the higher temperature at 6*0. Pionchon 

 also expresses the specific heat of iron by the following 

 formulae, in which 7 is the specific heat at the temperature 

 t°G. 



0° C. to 660° C. 7 =0-11012 + 0-00005066 1 + 0*000000164 1\ 

 660° to 720° y =0-57803-0-002872 1 + 0-000003585 1*. 

 720° to 1000° 7 =Q-218 

 1050° to 1200° 7=0-19887. 



The author now believes that the lower of the two critical 

 temperatures quoted by him previously, and deduced from his 

 own experiments on the internal friction of iron, may be a 

 third critical temperature J, and not, as he at first supposed, the 

 temperature at which iron begins to suddenly lose its mag- 

 netic properties, which last seems beyond a doubt to be 

 Pionchon's first critical temperature. 



In the researches of Moissan § on the allotropic modifica- 

 tions of iron oxide two temperatures are very conspicuous, 

 namely, 700°C. and 1000° C. 



The following table, compiled from the various researches 

 mentioned above, will show at a glance the different critical 

 temperatures of nickel and iron which the author believes 

 may exist || : — 



* Phil. Mag. Feb. 1888. 



t Ann. de Phys. et de Chim. t. xi. (1887). 



% If this be so, there is probably another sudden change in the internal 

 friction at 680° C. 



§ Ann. de Phys. et de Chim. t. xxi. 



|| The changes taking place at A may, however, be due to approach of 

 the critical temperature B. 



