24 BULLETIN 983, U. S. DEPARTMENT OF AGRICULTURE. 
A constant-temperature bath, which will regulate to ±0.05° C, 
also materially assisted in the accuracy of the alcohol determination. 
The beers were cooled to 20° C. Portions of 100 c. c. each were taken 
for distillation, to which were added 50 c. c. of distilled water and 
about 5 grams of precipitated calcium carbonate, along with 3 or 4 
drops of high-boiling paraffin oil to prevent foaming. A portion of 
100 c. c. was distilled over and caught in a volumetric flask. If a 
drop or two of paraffin oil came over, it was readily removed with a 
small strip of absorbent paper toweling. The 100 c. c. of distillates 
was then placed in the 15.6° C. constant-temperature bath; after it 
came to temperature the volume was made up, and the specific gravity 
was determined by means of a Boots double-wall vacuum pycnometer. 
In this way the alcohol-content of the beer could be determined very 
accurately. Although only a small amount of calcium carbonate was 
necessary to neutralize the volatile acid present, as can be shown by a 
redistillation of the first distillate, an excess of calcium carbonate — 
about 5 grams — was used to prevent bumping. 
Beginning with fermentation No. 10, the total solids in the neutral 
juice and beer were determined. This was done in an attempt to 
correlate the specific gravity, Brix, and sugar data in the neutral 
juice, and also to give a check on the fermentable sugars and alcohol 
determinations. It has been found, for instance, in the fermentation 
of waste sulphite liquors that sufficient volatile compounds, mostly 
sulphur compounds, distill over in the alcohol determination to make 
this determination from the gravity of the distillate practically worth- 
less. The addition of alkali and the redistillation helped, but even 
then there was not much correlation between the alcohol as deter- 
mined and the sugar data. When determinations were made of total 
solids, however, it was found that, if the difference in the two deter- 
minations before and after fermentation were assumed to be alcohol 
and carbon dioxide, and if the alcohol were calculated from that 
difference, the results agreed quite well with the other analytical data 
and especially with the yields of alcohol obtained commercially or on 
a large scale experimentally. The data obtained at the Forest 
Products Laboratory on total solids, however, have not been of such 
assistance, and frequently the total solids do not even follow the spe- 
cific gravity or Brix readings; much less do they give a good indica- 
tion of the alcohol yields, as the following table will show: 
