488 Dr. J. E. Mills on the 



5. Bromo-benzene showed divergences at 30° and at 160° 

 and 170°. I thought these errors due to the Biot formula 

 and to the vapour- density *. Later these errors were greatly 

 reduced by a recalculation o£ the Biot formula by Young, 

 and by the slight revisious made by Young in some of the 

 volumes of the saturated vapour. At 30° the value still 

 differs from the mean by more than 2 per cent. At this 

 temperature the vapour-pressure has a value of only 5*67 



millimetres of mercury, and the -^ is but a fraction of a 



millimetre. It can be easily seen that one could not hope to 

 obtain with any degree of certainty the true value (within 



2 per cent.) of the -^ from the Biot formula at this, its end 



point. The calculated vapour- density is also used, and this 

 introduces some error. 



6. Iodo-benzene. — I thought the divergences at 190° to 

 210° to be due to the vapour-density (volume) and so stated "|\ 

 Later Young's revisions showed these errors to be largely 

 due to the vapour-density but partly also to the Biot formula. 



7. Methyl formate. — In 1906 J I stated that the divergences 

 at 30° and 40° were probably due to the vapour-density. 

 Later, Young resmoothed these values and the divergences 

 disappeared. A change in the value of the mean likewise 

 caused the divergence at 180° to decrease to within the 

 2 per cent, limit. 



8. Ethyl acetate. — A change in the average due to the 

 revision of the vapour-density by Young caused the diver- 

 gences at 220° and 230° to come within the 2 per cent, limit. 



9. Methyl isobutyrate. — At 130° a misprint of 105 for 103 

 in the volume of the vapour was found by Dr. Young § upon 

 examining the original manuscript at my request, and a 

 corresponding " divergence " in the constant disappeared. 

 Other misprints were found in the course of my work, but 



1 shall always consider the discovery of this particular mis- 

 print an unusual achievement had it been accomplished by 

 the use of a merely " accidental 1 '' relation. Recalculation of 

 the Biot formula and resmoothing of the vapour-density by 

 Young also caused the divergence at 260° to come within the 



2 per cent, limit. 



10. The values at 0° C. were often high. This was known 

 all along to be due in part to the use of the calculated density 



* Journ. Phys. Chem. viii. p. 400 (1904). 



f Ibid. viii. p. 401 (1904). 



X Ibid. x. p. 18 (1906). 



§ Ibid. xi. pp. 621-622 (1907). 



