548 Mr. F. E. S.nith on Bridge Methods for Resistance 

 At the sulphur point the results were as follows : — 



Table VI. 

 Observations at the sulphur point. Thermometer K 5. 



i 

 (amperes). 



i 2 . 



b.w.r. 



o-oio 



00001 



9-19 



0-020 



0-0U04 



9-25 



030 



0-0009 



934 1 



0040 



00016 



9-46 



0-050 



0-0025 



9-64 



0-060 



0-0036 



9-84 



o-oio 



00001 



9-20 



Az' 2 = At— A b.w.r. 

 (i 2 - 0-0001). = (b.w.r. -9-19). 



0-0003 

 0-0008 

 0-0015 

 00024 

 0-0035 

 0-0035* 



0-06 

 0-15 

 0-27 

 045 

 065 

 0-61* 



Calculated 

 correction 

 = i 2 Xl85. 



o-oi 9 



07 4 

 0-JL6 7 



0-29 6 

 0-46 3 

 0-68 G 

 00L 



2A//SAi 2 = 2-22/0-0X20 = 1-85, i.e. at the sulphur point 0- L ampere 

 would necessitate the application of a correction of 

 T-85 c. 



* These values are (0-0036 -r) and (9'84 -b.w.r.) respectively. 



Summarising the results of Tables IV. to VI. we have for 

 this particular thermometer and bridge, corrections to the 

 temperature reading* of 1° # 62, 1°'68, and 1 0, 85 C, at tem- 

 peratures in the neighbourhood of 0°, 100°, and 445° respec- 

 tively for a measuring current of 0*1 ampere ; and corrections 

 of a 016, 0°-017, and 0°'019 for the usual current, viz., 

 0*01 ampere, employed in precise work. The corrections at 

 0° and 100° may be taken as identical with the value 0°-01G 

 given by Prof. Callendar in 1902 for the same type of 

 thermometer, and his suggestion to keep the current and 

 not the watts constant, at all temperatures, is evidently the 

 one for general adoption. If this rule be followed, an 

 adjustment of an external resistance is in general necessary 

 as the temperature changes. Thus in the usual Callendar- 

 (iriffiths bridge with 5-ohm ratio coils, a 1 ohm fundamental 

 interval thermometer, and an applied E.M.F. of 2 volts 

 from a storage-cell, the external resistance at 0° 0. would 

 be about 133 ohms, and at 100° C. 118 ohms to maintain a 

 constant current of 0*01 ampere through the thermometer. 

 However, if the resistance was not varied a maximum error 

 of 0°-004 only would result over the range 0° to 100°. 



