THE BRISTOL PERIOD 77 



substance, for interpolation and for extrapolation below 220, 

 and we found that below this temperature Regnault's pressures 

 were much too high and that the formula R' =R +c(t' -t) was 

 applicable to our results between 220 and 440, though not to 

 Regnault's. [Trans. Chem. Soc. 49, 37 (1886).] 



The two generalisations were found not to be applicable, as 

 a rule, to dissociating substances, nor, for a very wide range of 

 pressure, to compounds containing a hydroxyl group, such as 

 water, the alcohols and the fatty acids. For moderate ranges 

 of pressure, however, the deviations are hardly noticeable in the 

 case of the hydroxyl compounds, so that no serious harm has been 

 done by choosing water as the standard substance. 



Landolt's vapour pressures of the lower fatty acids were also 

 found to be inaccurate by means of the same formula and Dr. 

 Arthur Richardson afterwards redetermined these pressures in 

 the Bristol laboratory. 



I may mention that Regnault's boiling point of sulphur was 

 subsequently found by Callendar and Griffiths to be too high, 

 and the vapour pressures of mercury had therefore to be again 

 corrected. [Young, Trans. Chem. Soc. 59, 629 (1891).] Ramsay 

 accepted my recalculations. 



Our investigations of the vapour pressures and volatilising 

 and boiling points of solids and liquids were extended to dis- 

 sociating substances [' Evaporation and Dissociation/ Pt. I., 

 Phil. Trans. 177, 71 (1886)], of which we examined a considerable 

 number chloral hydrate, butyl-chloral hydrate, chloral methyl- 

 alcoholate chloral ethyl-alcoholate, ammonium carbamate, am- 

 monium chloride, phthalic acid, succinic acid, aldehyde ammonia 

 and nitrogen peroxide. A special form of apparatus was required 

 for nitrogen peroxide on account of its action on mercury, and a 

 special method of calculation for ammonium chloride which acts 

 on mercury at high temperatures. Vapour density determina- 

 tions were also carried out. 



In the case of nitrogen peroxide and ammonium chloride 



