190 Holborn and Day — Gas Thermometer at 



was so regulated that the slowly rising temperature could be 

 easily observed at any time. It was even possible to observe 

 the stationary temperature at the melting point just before the 

 interruption of the circuit. The observed melting tempera- 

 tures in microvolts and in degrees follow : 







Microvolts. 



Degrees. 





Oct. 17 (I) 



10209 



1064-5 







10209 



1064-5 







10206 



1064-2 







10208 



1064-4 





Oct. 18 (II) 



10202 



1063-9 







10201 



1063-8 





Nov. 15 (II) 



10211 



1064-6 







10211 



1064-6 







10211 



1064-6 





Jan. 2 (II) 



10190 



1062-9 







10190 



1062-9 







10200 



1063-7 







10189 



1062-8 





Jan. 3 (II) 



10203 



1064-0 







10197 



1063-5 







10213 



1064-8 



e mean 



is 1064-0° ± 0-6°. 





Silver — Wire Method. — The melting point of silver was 

 also first determined by the wire method and two specimens 

 were used, both from the Frankfurt Scheideanstalt. One of 

 these was in the form of wire 05 mm and 025 mm in diameter and the 

 other a piece of sheet silver 025 mm in thickness, from which thin 

 strips were cut and inserted in the junction of the element. 

 Here again no difference in the melting temperature of the 

 two specimens could be established. The same ovens were used 

 as for the gold determination. The results follow : 



Sept. 25 



Oct. 16 



Oct. IS 



Microvolts. 



Degrees. 



8928 



953-3° 



8936 



954-0 



8937 



954-1 



8922 



952-8 



8922 



952-8 



8914 



952-1 



8927 



953-3 



8930 



953-5 



8935 



953-9 



8939 



954-3 



8958 



956-0 



8925 



953-1 



8941 



954-5 



8924 



953-0 



Nov. 15 



The mean is 953*6° with a mean error of ± 0-9°. 



