Research on some Standards of Light. 



497 



conditions " (see Table V and fig. 8). It must, however, be borne in 

 mind, that for the final experiments it would be advisable to increase 

 the mass of metal to 1000 or 2000 grammes, and this would entail a 

 change in the quantity of gas burnt.* 



This said, let us proceed to determine to what extent the intensity of 

 the light is dependent on the experimental conditions. 



Change in the shape of the platinum ingot. — By changing the shape of 

 the ingot we shall modify not only its rate of cooling, but also the 

 maximum temperature it will reach under the " normal " supply of 

 gas. If we flatten out the disc beyond a certain limit, only the central 

 portion of the platinum will actually fuse. On the other hand, if we 

 make the superficial area too small, forming the metal into a long 

 cylinder, a limit will be reached when the fusion will cease to extend 

 to the lower layers. In the present case 35 sq, cm. and 8 sq. cm. are 



Table VII. (See fig. 11.) 



Superficial area of Platinum Ingot 10 sq. cm. 



Intensity of the Light from 1 sq. cm. of Platinum at the Temperature 

 of Solidification = 1-004. 



Time. 1 



Photometer 

 reading. 2 



Light. 3 



10 



141-8 



1-232 



20 



144-8 



1-144 



30 



148-8 



1-038 



40 



148-3 



1-050 



50 



149-6 



1018 



60 



150-3 



1 001 



70 



150 



1-008 



80 



150-1 



1-005 



90 



150 -2 



1-003 



100 



150 -2 



1 -003 



110 



151 -5 



0-972 



120 



152-8 



0-942 



130 



154 -0 



0-915 



140 



154-2 



0-910 



150 



164-8 



0-704 



160 



173 -3 



0-573 



1 The time is counted from the instant at which the blow-pipe was stopped. 



2 The photometer reading is the distance in centimetres between the photometer 

 head and the incandescent lamp which served as a standard of reference. 



3 The unit of light is the mean of a number of determinations made under the 

 " normal " conditions. 



* It would also be advisable in future experiments to shape the platinum ingot 

 into a half sphere, as this shape would be preferable with regard to the constancy 

 of the observations during the time the metal is freezing. 



