sosman: problems of the oxides of iron 63 



neighborhood of -40? The susceptibility rises to a maxi- 

 mum at about 600°, and then falls rapidly to about half value 

 at 690°, where the curve changes direction practically by a 

 right angle, the susceptibility being nearly constant from 690° 

 to 1300? 



We have confirmed the high temperature inversion, placing 

 it at 678° by means of thermal curves. 12 The inversion is sharp 

 and reproducible, and appears on both rising and falling tem- 

 perature curves. The —40° inversion we have not yet investi- 

 gated thermally. 



Curie 13 discovered a similar magnetic inversion in magnetite 

 at about 535? Barton and Williams 14 confirmed this inversion, 

 placing it about 545? Weiss and Foex 15 placed the inversion 

 at 581°, Wologdine 16 at 525? The temperature of the magnetic 

 inversion is undoubtedly influenced by the magnitude of the 

 induction or of the magnetizing force. No thermal study of 

 the inversion has yet been made. The principal precaution 

 to be observed is the avoidance of oxidation of the magnetite 

 by air. Some of the conclusions of Weiss and Foex seem to 

 have been vitiated by a slight oxidation of their artificial magnet- 

 ite, and a faint trace of the 678° point of Fe 2 3 appears on their 

 magnetite curve. 



The magnetic inversion in pure iron occurs at about 770°, 

 varying somewhat with the magnetizing force to which the 

 iron is subjected. It is found thermally at 768°, and by electrical 

 resistance, at 757? This is the thermal inversion known as A2, 

 whose existence in pure iron free from carbon was seriously 

 questioned some years ago, but which has been established as 

 an independent and characteristic inversion by the excellent 

 work of Burgess and his co-workers at the Bureau of Standards. 17 



These magnetic inversions, which occur not only in iron, mag- 

 netite, and hematite, but also in various ferrites, are all very 



12 Sosman and Hostetter. Jour. Am. Chem. Soc, 38: 831. 1916. 



13 Curie, P. Ann. chim. phys., 5: 322. 1895. 



14 Barton, E. H., and Williams, W. British Assoc. Rep., pp. 657-658. 1892. 



15 Weiss, P., and Foex, G. Arch. sci. phys. (Geneve), 31 : 5-19, 89-117. 1911. 



16 Compt. rend., 148: 776. 1909. 



17 Burgess and Crowe. Bull. Bur. Std., 10:315-370. 1914. Burgess and 

 Kellberg. Ibid., 11:457-470. 1914. ' 



