392 Bulletin 182. 



the crop where so few as two beets are used as a sample. It was, 

 therefore, decided that each sample should consist of six beets, 

 which were forwarded by express, instead of through the mails 

 as formerly, thus materially reducing the liability to error from 

 the cause mentioned. 



The season for preparing the land and sowing was unusually 

 favorable so that the seed was gotten in the soil in line condition 

 and at an early date. In fact the early date at which the seeding 

 was completed had the effect of reducing the number of our experi- 

 ments, as we liad expected at least two weeks more time in which 

 to go among tlie farmers to arrange the work. The weather condi- 

 tions were favorable for germination, a good stand was obtained and 

 early and economical tillage was generally facilitated so that up to 

 the middle of August the crop was unusually promising. From 

 this time on the effect of drought, which in some localities was the 

 most destructive on record, was verv manifest. In some instances 

 the damage was so great that the experiment was abandoned as not 

 likely to give trustworthy data. In fact it is believed that abnormal 

 conditions of any kind, and especially drought, lessen the value of 

 experimental data as they emphasize the unavoidable inequalities of 

 soil as regards texture, natural water supply, etc. 



The variety tests. — The following table. No. 1, gives a statement, 

 somewhat in detail, of the results obtained by each experimenter. 

 The table gives the name and address of the grower, the character 

 of the soil, the varieties of beets grown, the yield per acre in tons, 

 the percentage of sugar in the beets and the purity of the juice. 



All analyses so recorded in this ])ulletin were made by Mr. G. W, 

 Cavanaugh and Mr. A. L. Knisely, Assistant Chemists of the 

 Experiment Station. 



