498 Prof. A. Schuster on the Scale- Value, of 



There is a marked difference between the results of this 

 investigation and that deduced by Rowland * from the com- 

 parison of Joule's thermometers with his Baudin No. 6166. 

 Tables VII. (referring to the first of the above series of 

 comparisons) and VIII. (to the second) are intended to bring- 

 out this difference. 



The numbers entered into the different columns of Table VII. 

 are as follows : — 



Column I. T T or the reading on the Tonnelot thermo- 

 meter. 

 „ II. T T — T. or the corresponding difference in the 

 reading on the scale used by Joule and the 

 Tonnelot scale. 

 „ III. T N — T T , the correction to the Chappuis nitrogen- 

 scale as interpolated between the numbers 

 given in the table at the end of Guillaume's 

 Thermometrie. 

 „ IV. The calculated difference (T— T N ) between the 



Joule and Chappuis nitrogen-scale. 

 „ V. The corresponding difference (T.—T N ) E between 

 the Joule and Rowland's air-thermometers. 

 „ VI. The difference 8 between the numbers given in 

 Columns IV. and V. 

 A word of explanation is necessary as to how the numbers 

 of Column V. have been obtained. Rowland gives in his 

 paper the difference in the readings between Joule A and what 

 he calls the " perfect " air-thermometer at a great number of 

 points, none of them corresponding of course exactly to those 

 of Column L, for which they are here required. I have taken 

 the average between the value given for the temperature which 

 lies nearest to that of Column I. and the two which lie imme- 

 diately above and below it. The figures alter sufficiently 

 slowly and with sufficient regularity to allow us to consider 

 the numbers thus found as substantially correct. 



We may deal more simply with the numbers obtained in 

 the second series. The comparison between the Tonnelot and 

 the Baudin thermometer already referred to gave, for the 

 connexion between the two, the equation 



T T -T B = -0194- -00089 T B . 



Combining with this the experimental connexion between the 

 Joule and Baudin, 



T B -T / =--0017 + -00084T B , 



T T -T.=-0177--00005T T ; 

 * Proc. Anaer. Acad. xvi. p. 38. 



