ON STANDARDS OF LIGHT. 47 



Appendix II. 

 Incandescent Platinum. 



The attention of the Committee was naturally directed to the pro- 

 posed Frencli standard suggested by M. Violle, namely, the light 

 emitted by a square centimetre of liquid platinum at the solidifying 

 point. The apparatus required for the production of such a standard is 

 of necessity very cumbrous and inconvenient and extremely ill-suited for 

 photometrical measurements. The attention of the Committee was there- 

 fore directed to methods of obtaining the same result in a more con- 

 venient manner and on a smaller scale, with the view of constructing 

 apparatus which could without inconvenience be introduced into an 

 ordinary photometer. 



For this purpose the apparatus employed by Mr. Dibdin in his recent 

 experiments was first tried. This consists of an arrangement by means 

 of which a roll of platinum foil was stretched over two rollers, having an 

 interval of about three inches between them, across which the foU was 

 stretched. In front of the foil a steatite plate having a circular aperture 

 of a quarter of an inch was placed, and immediately behind the foil was 

 an oxyhydrogen burner so arranged that when a full flame was turned on 

 it impinged upon the foil in a dii'ection horizontal with the aperture in 

 the steatite plate and thus heated the platinum to fusion. Although the 

 mechanical part worked as satisfactorily as could be desired, the results 

 obtained on the photometer were so variable that no reliance could be 

 placed upon them. The method was then modified by placing the foil 

 between two plates of perforated steatite so as to inclose the portion to be 

 heated in a steatite cell. It was found, however, that the close contact of 

 the steatite conducted the heat away from the platinum so rapidly that 

 only an ii-regular portion in the middle of the cell was actually fused, and 

 the results were even more unsatisfactory than before. 



It was next considered desirable to vary the method by heatino- a 

 thick rod of platinum to its melting-point and allowing the light from 

 the fused bead of metal thus obtained to pass to the photometer disc 

 through a small opening in a suitable mask. This method was found to 

 be all but impracticable, but it gave results of value as indicatino- one 

 cause of the uncertainty in the quantity of light emitted from molten 

 metallic surfaces. It was observed that small scum-like particles were 

 constantly floating over the surface of the molten platinum, and that 

 these particles gave far more light than the mass of metal itself. Con- 

 sequently as there existed no means of ensuring the absence of such 

 particles it was obvious that there could be little certainty in the light 

 emitted from such unevenly illuminated surfaces. 



In addition to the blowpipe experiments, it appeared desirable to try 

 the experiment of fusing platinum by means of the electric current as it 

 seemed probable that by this means the platinum could be kept in any 

 given condition of incandescence for a longer period, and that the experi- 

 ment would be more under control. An apparatus for the purpose was 

 consequently arranged. A strip of platinum was held between two metal 

 •clips insulated from each other. Close in front of the platinum was 

 j)laced a small screen having a perforation of a quarter of an inch in 



