492 



Mr J. E. Petavel. An Experimental 



of temperature the cell was placed in the inner chamber of a calori- 

 meter. Under these circumstances the pressure on the terminals of 

 the lamp could be kept constant during many hours to within one 

 hundredth of a per cent. 



Table V. (See fig. 8.)— Normal Conditions. 



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

 of Solidification as deduced from the observations given below 

 is 1*002 (the error of this determination is therefore 0*002). 



Time. 



Photometer 

 reading. 



Light. 



10 



118 *3 



2*236 



20 



133*8 



1 -503 



30 



143 *0 



1*196 



40 



154*3 



0*908 



50 



164*5 



0-693 



60 



171*8 



0*623 



70 



174-4 



0-559 



80 



150 *0 



1 -008 



90 



150*3 



1*001 



100 



150 -6 



0*993 



110 



150*7 



0-991 



120 



150 -8 



0-988 



130 



150 -9 



0-986 



140 



151-0 



0-984 



150 



154-8 



0-897 



160 



163-8 



0-722 



Having in the manner indicated above obtained a set of photometric 

 readings at intervals of ten seconds from the time the blow-pipe was 

 stopped, let us consider what is the best manner of reducing these 

 observations. The typical shape of the curve obtained is shown in 

 fig. 8 (see Table V) ; the abscissae represent time, the ordinates light. 

 The curve consists of three parts, the first falling rapidly represents 

 the decrease of light sent out from the liquid metal as its temperature 

 falls. There is a sudden break in the curve when the platinum begins 

 to solidify, followed by a practically straight line. This second period 

 (forty to fifty seconds in the conditions under which these experiments 

 were made) marks the time during which the platinum ingot is freez- 

 ing. For the sake of brevity we shall in future refer to this part of 

 the curve as the "constant" period, though strictly speaking, owing 

 to the imperfect conduction of the metal, the intensity of the light 

 decreases slightly as the time increases. The third part of the curve 

 represents the cooling of the surface after the entire mass is solid. The 

 want of sharpness in the transition between the second and third parts 



