Pyrometer on the Melting-points of Gold and Silver. 225 



To facilitate observations, the apparatus was so arranged 

 that small changes of temperature could be followed and 

 recorded by watching the movements of a spot of light 

 reflected by the galvanometer-mirror on to a scale*. The 

 instrument was adjusted so that a deflexion of 40 milliin. on 

 the scale was equivalent to 10° Pt. 



The silver was melted in the usual way and well stirred. 

 The resistances in the box were adjusted approximately to the 

 balance. Readings of the position of the spot of light on the 

 scale were taken at intervals of 10 seconds as the mass cooled. 

 These readings were plotted in the form of curves, as shown 

 in Plate V., giving a continuous record of the cooling. 



The uppermost of the curves, No. 1, is the record of the 

 cooling of silver nearly free from oxygen. The temperature 

 fell rapidly to 830°, the normal freezing-point. It then 

 remained stationary to a tenth of a degree for about two 

 minutes and a half. The lower part of the silver having by 

 this time completely solidified while the upper part was still 

 partially liquid, the temperature began to fall, slowly at first, 

 and after about two minutes, when the whole mass was solid, 

 very rapidly. A little spitting was noticed between 810° and 

 805° Pt., but it was extremely slight and faint. 



The silver was then remelted and heated up to about 

 890° Pt. Oxygen was blown through it for about 30 or 40 

 seconds. This time the temperature fell to 826°, as shown 

 in curve No. 2, before it began to freeze. The tempera- 

 ture remained nearly stationary at first, falling less than half 

 a degree in the first minute. It then began to fall more 

 rapidly. Spitting commenced at about 814°, and became 

 very violent between 810° and 805°. It is evident from an 

 inspection of the curve that much heat was evolved during the 

 spitting, since the rate of cooling decreased considerably when 

 the spitting was at a maximum. Shortly after the spitting 

 ceased the temperature began to fall at the normal rate, as in 

 curve No. 1. 



The silver was then remelted with a reducing-flame, and 

 another record of cooling was taken without the addition of 

 oxygen. The curve proved to be an exact repetition of 

 No. 1. 



The silver was then heated to 870° Pt., and more oxvgen 

 passed for about a minute and a half. To my surprise, on 

 cooling, it began to freeze at 828°*8, as shown in curve 3, 



* For this purpose it was found necessary to compensate the galvano- 

 meter in such a way that its deflexions were independent of the tempe- 

 rature and resistance of its coils and of the strength of the battery. 



