Melting-points of Aluminium, Silver, Gold, Copper, fyc. 49 



The concordance of these results on various coppers, together 

 with the completely satisfactory behaviour of the metal in 

 fusion, and the ease and cheapness of obtaining the metal of 

 a very high grade of fineness, suggest the decided availability 

 of copper in a direct study of high temperatures or melting- 

 points by the gas-thermometer. A large mass of the metal 

 could be employed, and a constant and uniform temperature 

 for a protracted period thus secured for the bulb of the 

 gas-thermometer, or for other apparatus immersed in the 

 molten or solidifying material. There are unfortunately too 

 few substances which fulfil even these requirements. An 

 added merit lies in the nearness to the gold melting-point, 

 enabling the two to be satisfactorily connected by some means 

 of relative measurement. 



It also appears that the use of good commercial copper 

 would introduce sensibly less error (3° less) into the calibra- 

 tion of the Le Chatelier pyrometer than the use of the 

 " dentists' gold " above tested, which is as good metal as 

 would readily be obtained in the market by most observers. 

 Reliability of the Results. — The points involved are : — 

 Instrumental errors. 

 Purity of the metal. 



Was the observed point the real melting-point ? 

 Validity of the interpolation equation. 



Error in the assumed melting-point of gold and boiling- 

 point of sulphur. 



The investigation was planned and the apparatus arranged 

 with the intention of reducing the combined instrumental 

 errors below one-tenth of one per cent, in the measurement 

 of %e above 200° C. Tests, check measurements, and a dis- 

 cussion of the sources of error, unnecessary to detail here, 

 have given satisfactory demonstration that an even higher 

 accuracy than this was attained. So far, therefore, as constant 

 or variable instrumental errors are concerned, it is believed 

 that no error beyond o, 5 to 1° 0. exists in the results, while 

 probably this estimate is large. 



The error from impurities must have been exceptionally 

 small, as the analysis of the metals indicates. Some impurities 

 from alloying with the platinum and rhodium of the thermo- 

 couple must have entered during the experimenting, but 

 as results at different stages of the work checked those 

 obtained upon the first use of the metal, and as renewals of 

 the metal made no difference in readings beyond the limits of 

 other variations (about 5 parts in 10,000), the error from this 

 source must have been negligible. 



In the case of platinum the metal at command was un- 

 Phil. Mag. S. 5. Vol. 42. No. 254. July 1896. E 



