( 645 ) 



foil and iiidiariibber solution in one of the ways represented in 

 figs. 2, 3 and 5. 



For the filling of the tnbes we can recommend in general di-y 

 hydrogen. 



§ 2. Batterij of stuiubird-thermoelements. At the end of Comm. N". 27 

 it was remarked liiat the comparison-thermoelement itself with 

 junctnres placed in ice and steam, tiie electromotive force of the obser- 

 vation-element being e.\{)ressed in terms of that of the comparison- 

 thermoelement, conld Ite used as a standai'd. One of the junctures 

 is easily kejü at O*^ C. when the precautions are taken described in 

 Comm. N". 27 and in this pa])er §1. As to the other which is [jlaced 

 in steam we must make sure tiiat in tlie l)oiling-apparatus described 

 in Comm. N". 27 tiie water vapour flows out at a stead^M'ate and we 

 must apply tiie small correction for the variation of the boiling-point 

 with the barometric height in order to easily reduce the electromotive 

 force to that which would be obtained if this juncture were kept 

 precisely at J 00' C. If the metals do not undergo a secular variation, 

 we always have at our disposal a constant electromotive force 

 (although it cannot be reproduced with certainty independent of the 

 special apparatus) wliich is very appropriate for the calibration of 

 sensitive galvanometers in general, and has moreover the advantage 

 in thermoelectric determinations of temperature that it belongs to the 

 same size as the electromotive force to be measured. 



It seems that this idea has later been developed by the Phys. 

 Techn. Reichsanstalt ^). 



At the time of Comm. N". 27 (.June '96) the comparison thermo- 

 element was compared at intervals with a standai'd-element (then 

 Clark's). On the whole it satisfied the re((uiremeuts better tlian the 

 CLARK-cell, which had to be replaced repeatedly, while owing to the 

 improvements of the protection of thermoelements we had obtained 

 comparison-thermoelements which coidd serve unaltered for indefinite 

 time. The favourable experiences made with the German-silver- 

 copper-element in Comm. N". 27 led us to replace the single comparison- 

 or standard-thermoelement by a battery of 3 standard-thermoelements 

 each with its own boiling-apparatus and ils own ice-pot, which are 

 mounted insulated. Two of those elements /^, and 7^, are m.ade of con- 

 stantin-steel, the thiixl Q^ is the afore-mentioned (German silver-copper- 

 element with a 3 times smaller electromotive force than that of 7^» or 1\. 



1) Vgl. Zeitschr. f. Inslik. Bd. 17, p. 174, 1897. 



Idem Bd. 18, p. 183, 1898. 



Idem Bd. 22, p. 149, 1902. 



Idem Bd. 23, p. 174, 1903. 



