188 



Holborn and Day — Gas Thermometer at 



the melting metal (l cm from the bottom of the crucible) and was 

 clamped in position there. 



3. 



J Natural size. 



The figure shows the oven with a graphite crucible in posi- 

 tion and a second slightly smaller one with, a hole bored in the 

 bottom inverted over it as a cover. 



The porcelain crucibles were similar in size and form, l mm 

 thick and without cover. 



Inasmuch as the electric heating furnished a much more 

 uniform distribution of temperature than the gas oven and as 

 no systematic differences between the melting and freezing 

 temperatures were observed even with the most varied condi- 

 tions of current and quantity of metal, we did not attempt to stir 

 the metals in general. Only in one section of the observations 

 with silver an effort was made to expose the melted metal more 

 thoroughly to the air by stirring. 



In the preliminary experiments the coil was wound directly 

 upon a crucible of fire clay, but that was soon given up — first 



