348 PROCEEDINGS OF THE AMERICAN ACADEMY. 



When one considers that the carbon bisuljjhide was evidently deter- 

 mined very roughly, the agreement is an excellent one. Here, too, 

 we find the existence of two curves. The last measurement lies oa 

 the second curve when water is the precipitate, and not carbon bisul- 

 phide. As only one point on this curve was measured, it is impossible 

 to determine the constants even approximately. The object of this 

 investigation by Tuchschmidt and Follenius was to obtain a method 

 for determining the strength of aqueous alcohol quickly and easily. 

 Owing to the unpleasant properties of carbon bisulphide, their choice of 

 liquids was bad, though the method seems to me a good one. If one 

 were to make a complete table for benzol or chloroform and aqueous 

 alcohol at zero degrees, the rest would be simplicity itself. One would 

 take ten cubic centimeters of the alcohol to be tested and run m chloro- 

 form from a burette till saturated, when a glance at the table would 

 give the percentage composition of the alcohol. The method would 

 be quicker thau any except with a hydrometer, and more accurate than 

 that. An idea of the accuracy is given by the fact that at 20°, 5 c.c. 

 of 96% alcohol require about 20 c.c. chloroform, while the same amount 

 of 97.5% alcohol requires about 30 c.c. for saturation. For a weaker 

 alcohol the change for each per cent is much less ; but the measurements 

 can be made more accurately. 



I will now take up the measurements of PfeifFer * on the miscibility 

 of various esters with alcohol and water. The measurements were 

 not made under the most favorable circumstances. A known amount 

 of the ester was poured into a beaker, a definite quantity of alcohol 

 added, and water run in till the saturation point was reached. Nothing 

 was done to prevent evaporation, nothing, so far as is mentioned, to 

 keep a constant temperature, and there was no means of warming the 

 solution above the final temperature in order to insure that equilib- 

 rium had been reached. The necessity of this last had already been 

 pointed out by Duclaux,t who found that, on cooling a solution below 

 its saturation point, it clouded at once ; but on warming, the equilibrium 

 was reached much more slowly. Given these untoward conditions, 

 the comparative accuracy of the measurements is remarkable. While 

 Pfeiffer has given us series upon series of valuable measurements, 

 showing the increase of miscibility of esters and water in presence of 

 alcohol, he has curiously enough omitted all determinations of the misci- 

 bility of water and esters when no alcohol is present. This is still 



* Zeitschr. f. ph. Chem., IX. 469. 1892. 

 t Ann. chim. phys., [5.], VII. 264. 1876. 



