EXPERIMENT STATION BULLETINS. 185 



whereby the experimenter can tell whether it fairly represents the plot 

 wherein it j?rows or not. The introduction of this element of chance 

 in securing; representative samples seriously injures the value of tables 

 in which varieties are compared and all tables as well where yields 

 of suprar per acre are contrasted. In practical work at the factory also 

 the inability to secure samples known to fairly represent the load 

 throws donbt upon the whole system of paying for beets according to 

 the sugar they are supposed to contain. 



Omitting varieties 3, 11. and 21 which were not reported from all tests, 

 and taking the averages of the remainder of the twenty-four varieties 

 grown at the college, at Alma and on the Goodnoe farm, we find that the 

 duplicate samples taken at the college on October 7 gave an average test 

 of 15.5 and 15.6 or practically identical and this notwithstanding the 

 fact that individual samples of the same variety varied so widely as we 

 have just seen ; on October 23 the per cent of sugar in these varieties had 

 increased from 15.5 to 10.2 ; the records at Alma, as determined by analy- 

 sis at the college, were but 15.3 per cent, more than a per cent less than 

 the same varieties grown at the college; on the Goodnoe farm, on the 

 other hand, the average per cent of sugar in these same varieties was 

 16.8, higher than either at the college or at Alma. 



These findings are interesting especially when it is remembered that 

 the Goodnoe soil was almost a muck and lay so low that it was flooded 

 with water for the entire season except in spots large enough to furnish 

 samples for analysis, and on these spots the water was so close to the 

 surface that the beets were round and turnip-shaped rather than conical. 



The principal factor in determining the per cent of sugar in the crop 

 of beets is the variety chosen. The beet is an exceedingly highly-bred plant, 

 selected and bred for many generations to secure the greatest possible 

 amount of sugar in the roots. No matter how well cared for the crop 

 may be, if the seed is not from selected stocks, the crop will be low in 

 sugar. The farmer cannot materially alter the per cent of sugar in the 

 crop by any methods of culture if he puts the crop on any soil other than 

 undigested muck or does not use too nitrogenous fertilizers. The farmer 

 must depend upon the factory to furnish him the seed that will grow beets 

 rich in sugar, knowing that he cannot affect the beets materially in this 

 respect. 



Further work will be needed to determine whether there is a material 

 difference in varieties as to their adaj)tation to different soils and condi- 

 tions. The season was too wet this year to make the above table of great 

 value as far as it bears on this question. 



The yields of the several varieties are reported in the next table. As 

 stated there were no yields computed from the Goodnoe plots. 



D. Spraying to Prevent Disease. 



In co-operation with the Department of Agriculture at Washington, 

 a series of experiments were undertaken to determine the practicability 

 of spraying \\nth. Bordeaux mixture to prevent leaf blight and possibly 

 other diseases of beets. The too frequent rains ruined the experiments and 

 there are no results to report. Leaf blight was very prevalent and undoubt- 

 edly lessened the yield of all plots reported in this Bulletin. It was 

 noted that the disease. was much less severe where beets did not follow 

 beets and also where the ground had been well fertilized with ashes, 

 the strong vigorous plants possibly being better able to resist. 

 24 



