BURGESS AND WALTENBERGI MELTING POINTS 375 



Kahlbaum and on one from Dr. H. Goldschmidt, all of a purity 

 probably not greater than 98 per cent. 



TABLE III . 



Melting Point of (98 per cent?) Chromium by Micropyrometer 



Kahlbaum I 1527° 



Kahlbaum II 1536°, 1513°, 1513° 



Goldschmidt 1514°, 1524° 



As the best representative value we may take 1520°. The 

 melting point of pure chromium, however, may well be above that 

 of iron, as may be shown by a consideration of the well known 

 formula for lowering the freezing point by metallic impurities. 

 On the other hand, the value 1520° may be more nearly correct 

 if the oxides present do not lower but raise the chromium melting 

 point as was found in the case of vanadium by Ruff and Martin.^ 

 If chromium is melted, for example, in a slightly oxidizing atmos- 

 phere or in impure hydrogen, an apparent melting point above 

 that of platinum may be obtained. 



Manganese. We have not been able to obtain pure manganese, 

 and the impure samples from Kahlbaum and from Goldschmidt 

 are even more sluggish than chromium. The Goldschmidt man- 

 ganese was from a sample of about 97.5 per cent pure. Melts 

 were taken both on platinum and on nickel strips with the micro- 

 pyrometer using gold (1063°) and palladium as calibration points. 

 Those on nickel were less satisfactory, due apparently to evapo- 

 ration of nickel. 



TABLE IV 



Melting Point of (97.51) Manganese by Micropyrometer 



Kahlbaum on Pt., 1242°; on Ni 1221° 



Goldschmidt on Pt., 1261°, 1264°, 1279°; on Ni 1254° 



The mean is 1254°, which is probably lower than the melting 

 point of strictly pure manganese by 20°. 



Vanadimyi. Dr. v. Wartenberg of Berlin kindly sent us some 

 97 per cent vanadium prepared by him from the suboxide by 



' Ruff and Martin, tJber reines Vanadin. Zs. Angew. Chem., 25: 49. 1912. 



