150 Professor Beck's Researches on Wines 



proportion of subcarbonate of potassa. These experiments seem 

 to prove, that the separation of alcohol in wine, by the common 

 mode of adding subcarbonate of potassa, is prevented by the 

 tartrate of lime, which is liberated by the first addition of an 

 alkali, and which, perhaps, together with some of the other mat- 

 ters, forms a flocculent mass suspended in the liquor. In my 

 opinion, they constitute a more decisive proof of the existence 

 of ready formed alcohol in wine, tlian any which has yet been 

 offered. 



In the analysis of wines and other fermented liquors, the re- 

 sults of which are given below, the following process was adopt- 

 ed, for the purpose of determining the proportion of alcohol 

 which they contain. 



A glass bottle, with a long and narrow neck, capable of hold- 

 ing 1020 grains of distilled water, was filled with the wine un- 

 der examination. This quantity of wine was now put into a 

 glass retort, which was carefully luted to a receiver, so as to 

 prevent the escape of vapour. A gentle heat was applied to the 

 retort, while the receiver was kept constantly cold, by the drop- 

 ping of water upon it. The heat being cautiously managed, to- 

 wards the end of the process I was enabled to distill off nearly 

 the whole of the wine, without burning the solid residuum. 

 Thus, from 1008 grains of Madeira, I distilled over 979 grains, 

 while from the Port there was a loss of nearly 50 grains, in the 

 same measure, owing chiefly to the greater proportion of resi- 

 duary matter. To make up for this loss, I added wine, of the 

 same kind as that distilled, so as to bring it to the original measure 

 of the wine, excepting a small allowance for the space occupied by 

 the solid ingredients. These were now shaken together, and al- 

 lowed to remain for some time, and the specific gravity then care- 

 fully determined at the temperature of 60" Fahr. From this the 

 proportions of alcohol and water by zcelg-ht, were ascertained by 

 a reference to the Tables of Mr Gilpin, published in the Philo- 

 sophical Transactions for 1794, and, by calculation, the propor- 

 tions by measure were estimated. 



This process, it will be observed, is essentially that of Mr 

 Brande, and that my results may be easily compared with his, 

 I have referred them to the same standard which he adopted. 

 I may also state, that I compared these results in one or two 



