452 A. L. Day and J. K. Clement — Gas Thermometer. 



ever before the desirability of some provision, preferably by the 

 'Bureau of Standards, for standard metals, the uniform purity 

 of which can be absolutely depended upon, in terms of which 

 such constants can be expressed. In the absence of such a 

 provision, it is difficult to see just how to make the gas scale 

 conveniently available for general use in its full accuracy. 

 This is furthermore a matter of considerable importance in 

 view of the extended extrapolation to which the gas scale is 

 frequently subjected by the use of thermoelements or other- 

 wise. Supposing the metal melting points to be capable of 

 reproducing the temperature curve correct within 1° at the 

 copper point (1081°), the extrapolation to 1500° may easily 

 remain uncertain by at least 5° in the hands of different indi- 

 viduals using the same functions for the extrapolation. 

 . JSTor is this the only difficulty to which this use of standard 

 melting points may lead. On account of the systematic errors 

 attending the application of an equation of the second degree 

 to the gas thermometer observations, to which reference has 

 been made (p. 448), the usual standard melting points are not 

 competent to reproduce the gas scale exactly. A curve of the 

 second degree, developed by least square solution from seventy- 

 six observations which show systematic deviation, cannot 

 be reproduced with only the three observations which chance 

 to represent the standard melting points. This will be imme- 

 diately apparent, though on a somewhat exaggerated scale, 

 if we locate on the dotted curve in fig. 8 the tempera- 

 tures corresponding to the zinc (419'5°), silver (957*9°) and 

 copper (1080-9°) points and then undertake to reproduce the 

 46 calculated" curve from them alone. 



As this is the method almost universally used for the pur- 

 pose, it is worth while indicating by a special case exactly 

 where it leads. Column 3 of the table below contains the 

 actual melting points of four of the purest metals interpolated 

 from gas thermometer observations close by. Column 4 

 contains the same points computed from Equation I. If 



Table V. 



Zinc 



Silver.... 

 Gold ...: 



Copper . . 



Element 



3403' 



9046' 



10214' 



10461' 



Temperatures 

 extrapolated 

 from nearest 

 observation 



418-48 



957-90 



1059*26 



1080-92 



Temperatures 



Calculated 



I 



418-87 



958*50 



1059-42 



1080-29 



Difference 



— •39 



— •60 



— •16 



+ •63 



