486 



Mr. J. E. Petavel. An Experimental 



oxy-hydrogen flame is used to fuse the platinum, but, taking every- 

 thing into account, the electrical method, if carried out on a sufficient 

 scale, would not only be more simple, but would afford a more ready 

 means of varying the rate of cooling. 



On the Fusion of Platinum by the Oxy-hydrogen Bloiv-joipe. 



The object of the following researches was to repeat the work done 

 some years ago by Violle, and to determine what are the best 

 experimental conditions. It is a somewhat curious fact that, although 

 Violle makes no mention of having encountered any special difficulties, 

 his experiments have never been repeated with success. Even under 

 the best conditions (such as are obtainable, for instance, at the 

 Reichsanstalt) the results have not proved satisfactory. Under these 

 circumstances, it seemed hardly advisable to attempt any further 

 research in this direction. But the theoretical advantages of the 

 Violle standard appeal strongly to anyone who has studied the 

 standards of light now in practical use, and it was thought that, 

 whatever the final result, the time spent in this study would not 

 be lost. 



The preliminary experiments occupied some considerable time, and 

 nearly six months passed before the conditions under which the fusion 

 should take place were clearly established. The platinum was fused 

 several hundred times, the conditions being varied in every conceivable 

 way. The shape of the furnace, the material of which it was made, 

 the form of the blow-pipe, the relative proportion of the gases, in fact, 

 everything that could have any bearing on the final result was in turn 

 studied, and the most favourable conditions determined. The slightest 

 impurity on the surface of the platinum has a considerable effect on the 

 quantity of light emitted. When cold the surface always appears 

 fairly bright and clean, but there is a temperature between a white and 

 red heat at which the smallest impurity is clearly visible. At this 

 temperature the platinum ingot, if it be quite pure, is very similar to 

 a pool of molten glass. Any small impurity will cause a slight haze, 

 resembling that produced on a mirror by a breath of moist air ; if the 

 impurities are present in a larger proportion the surface becomes very 

 similar to a sheet of ground glass. 



The necessary conditions to ensure a pure surface may be briefly 

 summed up as follows : — 



1. The platinum must be chemically pure. 



2. The crucible must be made of pure lime. 



Pure magnesia does not form a sufficiently coherent mass, and, alu- 

 mina being light and flocculent, allows the metal to sink through it. 



The lime should be entirely free from silica, one-quarter per cent, 

 being sufficient to spoil the surface. 



