19 



1883, 1 - ribed u method of driving out the volatile acids with steam, and this forms 

 tin- ba.-i.- "i the utlicial methods which for some years have been prescribed in Europe 

 and in Anu-rica. In essential details the present official methods appear to comply 

 with the following procedure: 



Fifty cubic centimeters of wine are distilled in a current of steam, in the meantime 

 heating the flask containing the sample until the liquid boils, and regulating the 

 flame so that the volume remains constant. Two hundred cubic centimeters of 

 distillate UP- collected and titrated with tenth-normal sodium hydroxid, using phenol- 

 phthalcin as indicator. 



< 'heini.-i s who have had considerable experience with this method must have noticed 

 that it often happens that distillates collected beyond 200 cc show a more or less 

 acid reaction. This is the case especially with certain red wines, notably 

 buivundys. prts, and clarets, and wines of the sauterne type. In fact, as the accom- 

 panying table .-how-, it .-eld "in if ever occurs that the first 200 cc distillate contains 

 even a fair approximation of the total volatile acids. In some instances it is seen that 

 by canying the distillation beyond L'OO cc the error is very considerable. Forexam- 

 ple. Pillowing closely the pp.vi.-ional method and carrying the distillation to 400 cc, 

 there are .-hown the following rates of increase in total volatile acids: In a burgundy, 

 approximately 1 1 per cent; in two samples of port, respectively, 13 and 14 per cent; 

 in a laret, 1:5 percent; and in a white wine, <) percent. There occur, indeed, wines 

 in which the volatil ;ear never to become completely exhausted, in which, 



in fact, i he di.-tillalr fails to appear permanently neutral even after very prolonged 

 di.-tillinu'. This phenomenon may be attributed not so much to acids of difficult 

 volatility as to a possible , i :: ion of the extract constituents under the influence 



longed heating by the direct action "f the flame which the official directions 

 ."iilil be maintained below the flask containing the sample. It would 

 t at ni"-t KM) cc, of distillate should contain practically 



all "f the volatile acid-, and that it may not be necessary or practical to prolong the 

 dUill.uion until the la.-t portion- of the distillate are neutral. 



IV ; .// th, 



a First 11 samples furnished by California Wine Association, 1906. 

 b Last 5 samples from miscellaneous sources. 



The results obtained on these 16 samples of wine show that in only four cases did 

 the fifth 100 cc distillate require as much as 0.2 cc of tenth-normal alkali to neutralize; 

 hence, f -r practical purposes, it may be assumed that the vanishing point of the 



U. S. Dept. Agr., Bureau of Chemistry, Bui. 107, p. 86. 



