93 



juice pressed out. The specific gravity of this was then observed, and the clarified 

 juice polarized. At harvesting time two or three different samples of each variety 

 were taken, and the results averaged. The sugar in the beets was determined in these 

 samples by the alcohol method of Tollens-Rapp-Degener (Koenig, Unters. landw. 

 wicht. Stoffe, 1891, p. 436). The results of the analyses are given in the following 

 table : 



Sugar beet season, 1801. 



SAMPLES TAKEN SEPTEMBER 26. 



SAMPLES TAKEN AT HARVESTING TIME, OCTOBER 26. 



The analyses of the samples taken September 26 agree as well as could be expected 

 with those of the samples taken at harvesting time. The latter samples were taken 

 from the harvested beets when a good idea could be obtained of the average size of 

 each variety. It may be said, in general, that the quality of the beets did not improve 

 after September 26, and it is not likely that the yield was increased perceptibly dur- 

 ing the month of October, owing to the extreme dryness of the soil. The beets were 

 very small, averaging only about 11 ounces for all the varieties. The average per cent 

 of sugar (sucrose) in the juice at harvesting time was 17.83 per cent, ranging from 

 14.99 to 20.53 per cent. The average sucrose in the beets was 15.50 per cent, with 

 13.27 per cent and 17.56 per cent as lowest and highest limit. By dividing 15.50 by 

 17.83 we find that the beets contained 86.9 per cent of juice on an average, showing 

 that the dry season produced beets with unnaturally high sugar content and with a 

 low percentage of juice. 



It will be noticed that the percentages of sugar increase as we go down in the 

 table that is, with the beets growing farther east on the plot. We saw that the soil 

 was drier and perhaps also poorer in the eastern part of the field than in the west- 

 ern, and the beets were smaller in size and richer in sugar the farther east we go in 

 the field. As a rule, size and sugar content of the beets stand in inverse ratio to one 

 another. 



