( 336 ) 



Pt"d were placed in the cryostat. Thus we simultaneously obtained 

 the results relating to them of the tables I and 11. Those of' the 

 second part of table II are derived from a comparison between Ft" a 

 and T 4 . The resistances observed were reduced with the factor 1.00073 

 (cf. the footnote on the preceding page), giving 18.560 and 18.398 i2. 

 These have been reduced with the data of table I to Pt'j. This 

 yields the values of the third column. 



The calibration of 7\ agrees satisfactorily with that of Bin- The 

 mean of the deviations for the two former and the two latter data 

 differs from table II by 0.022 <2 which corresponds to 0°.038. This 

 small difference between the readings of two entirely different gas 

 thermometers satisfactorily confirms the exactness of the limit of 

 accuracy derived formerly Comm. X". 95 c ) and enhances the relia- 

 bility of the other preceding determinations of temperature. 



!) The hydrogen thermometer temperatures are calculated on the scale of Comm. 

 N°. 95* (Oct. '06). As the zero pressure amounted to about 1360 mm. we have 

 accordingly accepted for the coeificient of pressure variation 0.0036629, derived 

 from the coefficient of pressure variation of Comm. N". 60 (Sept. 1900) (0.0036627 

 for 1100 mm. zero pressure). We have also applied a correction according to the 

 difference of the correction to the absolute scale for 1360 and 1100 mm. zero 

 pressure. 



