BANCROFT. — TERNARY MIXTURES. 349 



more remarkable if one considers the uselessness of comparing the effect 

 of equal quantities of alcohol on ethylaoetate and water, amylacetate and 

 water, without allowing for the fact that ethylacetate is roughly forty 

 times as soluble in water as amylacetate. As these solubilities had to 

 be known at any rate approximately, in order to apply Equation II. 

 to Pfeiffer's experimental data, I have determined several myself. 

 Through the courtesy of Mr. Dunlap of the organic laboratory, I re- 

 ceived small quantities of ethylbutyrate, ethylisovalerate, and isoamyl- 

 acetate. I dried them over calcium chloride and fractionated. The 

 change of boiling point of the portions used was four degrees for 

 the ethylisovalerate and two degrees for each of the others. The 

 amounts at my disposal made it not worth while to attempt further 

 purification. In Table XVI. I give the solubilities in cubic centimeters 

 of the solute in ten cubic centimeters of the solvent at 20°. For pur- 

 poses of comparison I have also inserted in this table the values for 

 ethylacetate from Table VIII. 



For these four esters I have found that the solubility in water 

 decreases with increasing temperature, while the solubility of water in 

 the ester increases with increasing temperature, both observations 

 being made at 20°. As there is no obvious reason why these four 

 esters should all be abnormal, it is more than likely that this behavior 

 is characteristic of all esters at ordinary temperatures. As it is im- 

 probable that the solubility of the esters in water can continue to 

 decrease indefinitely with increasing temperature, there must be some 

 point where it reaches a minimum, and it is quite possible that a deter- 

 mination of this temperature for different esters might give interesting 

 results. The experiments could be made with great ease as the 

 amount of saponification during the time necessary for a measurement 

 would be very small. 



