Practical Thcrmometric Standard. 531 



indicating apparatus, which is a matter of great convenience. 

 The instruments are also capable of continuous reading and 

 recording under any conditions, and are directly applicable to 

 all kinds of temperature measurement. The tendency at 

 present in France and Germany seems to be in the direction 

 of adopting the thermocouple as a practical standard for high 

 temperatures only, and the mercury-thermometer for the lower 

 parts of the scale ; but those who have used the platinum- 

 resistance thermometer would have no hesitation in preferring 

 it as a standard on account of the far greater accuracy 

 attainable over the whole range. I have already discussed 

 this point in a communication to the Phil. Mag. (Feb. 1*99, 

 p. 210), but it may be instructive to consider the question 

 from the point of view of the conditions above laid down. 



(II.) The most important qualification of a practical standard 

 is the possible accuracy of comparison and verification. 

 According to my experience, the order of consistency of 

 measurement attainable with a suitable platinum resistance- 

 thermometer is at least ten times that attainable with any 

 other kind of instrument at any part of the platinum range. 

 In order to take full advantage of this, and to avoid the small 

 differences between different platinum-thermometers due to 

 possible slight impurities in the platinum, it is only necessary 

 to select a particular standard platinum, as recommended in 

 the first of the proposals already submitted. The chief 

 objection to the adoption of the thermocouple as a standard, 

 from this point of view, is that, owing to the smallness of 

 the thermo-E.M.F-, it is impossible to attain sufficient delicacy 

 of reading, especially at moderate temperatures, where a 

 high order of accuracy is most important. 



With regard to portability and ease of reproduction, it is 

 sufficient to send a few grammes of the standard wire in an 

 ordinary letter, to reproduce the scale with the utmost 

 accuracy in any part of the world. 



(III.) The question of the difference between the practical 

 standard and the theoretical scale, and the facility of making 

 the comparison, is one of great interest. The order of 

 accuracy is here limited by that of the gas-thermometer; and 

 the relation between the scales will become more accurately 

 known as the latter instrument is improved. So far as can 

 at present be determined, it appears that the deviation of the 

 scale of the platinum resistance-thermometer from the theo- 

 retical scale is represented, almost within the limits of error 

 of the gas-thermometer, by a formula of the simplest possible 

 type. The verification of this formula at moderate tempera- 

 fcnr<0"» from 0° to 600 c (J., the most important part of the range, 



