

Manure should be used from one stable only, so that it is 

 the product of the same animals for a prolonged period. Never 

 take horse manure from livery stables, or mix sheep and cattle 

 manure. If possible, use the same pile of manure for the test 

 field and see that it is most carefully and evenly distributed 

 over the land in a uniform layer. Spread the manure always 

 under a right angle to the direction in which you intend to 

 plant your beet rows; it must be spread immediately and not 

 allowed to lie in piles, for even a day, in the field; it must be 

 plowed under as fast as it can be scattered. 



3rd: In spite of the most careful selection and prepara- 

 tion of the test field, there may be spots which at the first 

 glance plainly show that they differ in character from the gen- 

 eral make-up of the field. Such places should not be included in 

 the test at the time of harvesting. The beet rows should run 

 in right angles to the plow furrow and the other preparatory 

 operations, and parallel with the slope of the land. In case 

 the character of the soil or its depth change in a given direc- 

 tion, the beets should be planted in the same direction. 



4th: Even the most careful examination of the field and 

 the greatest care in its selection do not always guarantee that 

 conditions are absolutely uniform or will continue so during 

 the growing season. Heavy rains, accidents during the irriga- 

 tion, parasitic destruction and other agencies may cause great 

 variations in the final results. 



In order to eliminate the effects of such unavoidable occur- 

 rences as much as possible, and thus reduce to the lowest mar- 

 gin the errors resulting therefrom, each test must be repeated 

 several times ; and the more varieties are in the test the of tener 

 the rows should be repeated. There should never be less than 

 three series of each planting; one series of test rows is of no 

 value, two of very little. 



In order to obtain the most accurate results and data 

 which can be directly compared, the use of one seed variety, 

 as a ''standard" is highly recommended. This standard vari- 

 ety is always seeded alongside every other Variety included in 

 the test, and the returns of the latter are then directly com- 

 pared with the returns of the two "standard" varieties grow- 

 ing on either side. When a large number of varieties are to 



