Measurements of Precision in Platinum Thermometry. 543 



Compensating Leads. 



Three pairs of leads were examined. One pair, which had 

 been in use for five years, was obtained from Dr. Harker of 

 the Thermometric Department of the Laboratory, and two 

 new pairs were kindly lent to us by The Cambridge Scientific 

 Instrument Co. 



The leads obtained from Dr. Harker were of the usual 

 plaited flexible type, and were about 4 yards long. The 

 C leads had a resistance of about OOG ohm, and the P leads 

 were nominally equal in resistance to them. Resistance 

 measurements with an accuracy within 1 part in 10,000 

 showed that the resistance of the leads altered if they were 

 much disturbed. The ratio C/P was first measured as 

 1*0060, but we found during the first few minutes of our 

 work that the ratio was far from constant. 



In order to measure the changes with accuracy, the C and 

 P leads were connected so as to form two arms of a Wheat- 

 stone bridge, the other arms of which consisted of two 1-ohm 

 coils of manganin, which could be shunted to secure a 

 balance. The contacts were carefully made, and the 

 terminals and parts of the leads near thereto were guarded 

 against strain. The values of the ratio C/P given in the 

 first columns of Tables I. to III. were obtained when the 

 leads were stretched nearly to their full extent. When the 

 leads were still stretched but considerably agitated before 

 making the measurements the ratios in the second columns 

 were obtained, while those in the third columns relate to 

 observations made with the leads in the form of coils. There 

 does not appear to be any relation between the ratios and 

 the disposition of the leads. The ratios for any one pair of 

 leads were all measured on one day. 



Table I. 

 Leads from Dr. J. A. Harker. 

 Resistance of leads = about 060 ohm. 



„ P „ = „ 0-060 „ 

 Ratio C/P :— 



10059 



1-0049 



1-0050 



82 



43 



77 



99 



50 



42 



46 



58 





44 



67 

 51 



58 



1 



Maximum variation = 0'57 per cent. 



