ON THE SPECIFIC HEAT OF WATER. 



237 



as that which a formula of the third degree gives. Any formula of the third degree 



which fits the curve from C. to 

 70 C. tends to give too much drop 

 in the values in the neighlxmrhood 

 of 80 C., as may be well seen with 

 the curve which we have calculated 

 for BARNES' figures, and which is 

 given in the next section. 



15. Comparison of Results ivith 

 Former Determinations. Some of 

 the results of former observers are 

 given only relatively, and others in 

 absolute values. Relative figures 

 for the specific heat of water at 

 various temperatures referred to 

 the value at 15 C. as unity, and 

 taken with reference to the scale 

 of the hydrogen thermometer, are 

 set out in the ' Landolt-Bornstein 

 Tables ' (ed. 1905, p, 393). BARNES' 

 latest absolute values corrected for 

 the later value of the Clark cell 

 were not published till July, 1909 

 (' Roy. Soc. Proc.,' Ser. A, vol. 82, 

 p. 390), but the relative values 

 remain the same as given in the 

 above tables in 1905. In Table XII. 

 all these figures are set out, together 

 with our own figures obtained by 

 dividing our absolute values by our 

 absolute value for the 15 calorie. 

 The differences between our own 

 figures and the others are also 

 shown in the table. In fig. 10 all 

 these results are set out, together 

 also with REGNAULT'S figures, which, 

 though calculated for the air ther- 

 mometer scale and therefore not 

 entirely comparable, still afford a 

 useful basis for comparison when 



such large divergencies exist. As regards the general course of the curve, it will 



