224 Mr. Callendar on Experiments with a Platinum 



Le Chatelier has adopted for the graduation of his thermo- 

 couples the value 945° C. 



I have myself made some direct experiments on the freezing- 

 point of silver by means of air-thermometers with small bulbs, 

 about 2 cubic centim. capacity, both of platinum and por- 

 celain. These experiments point to a similar result, but 

 cannot be considered very satisfactory. The platinum oc- 

 cludes gas, and the glaze of the porcelain cracks so that the 

 material becomes porous. I hope shortly to be able to make 

 some bulbs of fused silica, which is a far more refractory 

 material, and ought to give better results. 



If we take for the present the value 945° C, and assume 

 that the coefficient Sin formula (d) is of the form (a + bt), 

 we find for this pyrometer a = 2*050, b=— "00065. Assu- 

 ming these values, we find for the freezing-point of gold the 

 figured 1037° C. 



This is lower than the usually accepted figure 1045° C, 

 which rests mainly on the experiments of Violle ; but if, as 

 Roberts- Austen thinks probable*, Violle's determinations of 

 the specific heat of gold are low, the figure he (Violle) 

 obtained for the melting-point of gold would be too high. 

 The value £=1037° C, given above, may therefore prove to 

 be the more correct. 



The lowering of the freezing-point of the silver by the 

 addition of 0*5 per cent, of lead was found to be 2°*3 on the 

 platinum scale. This becomes 2°*9 when reduced to the air- 

 thermometer scale. 



The alloy was subsequently assayed at the Mint and proved 

 to contain 0'54 per cent, of lead. This corresponds to 0'26 

 atoms of lead to 100 of silver. The fall produced per atom is 

 therefore 11°*2. Assuming the latent heat of fusion of silver 

 to be 21*07 (Person), the theoretical fall should be 12°-8. 

 It appears that the composition of the alloy was not uniform, 

 the upper portions being richer in lead. The agreement with 

 the theoretical fall is therefore quite as close as could be 

 expected. It may be noted that Heycock and Neville find 

 that silver dissolved in lead lowers the freezing-point exactly 

 to the extent predicted by theory on the assumption that its 

 molecule is monatomic in solution. 



I have since made some further experiments by a new 

 method with a view to investigate more particularly the con- 

 ditions under which silver absorbs oxygen, the extent to which 

 its freezing-point is thereby lowered, and the temperature at 

 which spitting occurs. 



* Proc. Roy. Soc. March 1891, p. 352. 



