122 Scientific Proceedings^ Royal Dublin Society. 



values approximates to a straightl-ine, but with formic acid there is a 

 deviation (only apparent on a large-scale drawing). 



This variation in the distillation constant may be due to several causes, 



(a) Variation in experimental conditions, such as irregular rate of 

 distillation, condensation in the still-head, &c. 



(&) Impurities in the acids employed. 



(c) Changes in the volumes of dilute solutions of tlie acids on further 

 dilution. 



{d) Deviation from the distribution law. 



The variations due to irregular condensation are reduced to a minimum 

 by surrounding the still-head with a steam-jacket, so that the temperature 

 is maintained at 100°. If the still-head were below 100°, condensation would 

 occur until the temperature had increased sufficiently for an equilibrium to 

 exist between the liquid and the vapour. When the temperature of the 

 still-head is above the boiling-point of tlie solvent in the flask, drops of the 

 solution which may have been carried into the neck of the flask become 

 completely volatilized, and in consequence an error is introduced. These 

 effects will not occur if the temperature of the neck is kept at the boiling- 

 point of the solvent used in the flask. Heating the flask by means of an 

 electric hot-plate secures a regular rate of ebullition, and therefore the 

 distillation will be regular if condensation is absent. 



Care was taken in preparing the acids for this investigation to obtain 

 them in as high a state of purity as possible. It is probable that the 

 formic, acetic, propionic, ii-butyric, and w-valeric acids used by the authors 

 contained no more than traces of impurities. 



The changes in volume on diluting the distillate, consisting of a dilute 

 solution of a fatty acid, with a solution of different concentration, are 

 probably negligible. A slight error, however, is introduced in assuming 

 that the weight of the distillate is equal to the weight of water contained in 

 it. Although the error is slight, it does not account for the alteration of the 

 constants with dilution. The effect is more marked in the case of formic 

 and acetic acids, where the error (due to assuming that the weight of distillate 

 is equal to the weight of water contained in it) is less than in the case of the 

 higher acids. 



In deducing expressions for the interpretation of the experimental results 

 it was assumed that there was no deviation from Nernst's law with change 

 in concentration. This assumption is not strictly justifiable, since changes 

 in molecular aggregation occur on diluting certain solutions. 



