6 



5 baskets, a portion may be discarded. It is essential, however, that 

 all baskets contain equal amounts of soil. If the amounts contained 

 in the baskets are unequal, the plants in those containing the largest 

 amounts will make the largest growth, just as those plants which have 

 ample room under field conditions grow larger than those that are 

 crowded. With this in* view it is desirable to ascertain the weight of 

 soil required to fill a basket to the desired fullness, then place an equal 

 weight of soil in each basket. Where distilled water is not available 

 for wetting the soils, clean, fresh rain water may be used. Well water 

 should be avoided, as it contains considerable soluble material that 

 might influence the growth of plants, and thus interfere with the 

 results. 



Experience has shown that a number of precautions are necessary 

 in order to secure reliable results in this work. In selecting samples 

 great care should be exercised in order to obtain results typical of the 

 field conditions. From 10 to 20 separate samples should be taken, 

 each of the same size and to the same depth, and distributed in differ- 

 ent parts of the field. The samples to be tested should be a composite 

 formed by a thorough mixture of these. 



Great care should also be exercised in selecting seed of the same size 

 and in the same stage of germination at the time of planting. Careful 

 mulching and paraffining of the baskets is essential to secure reliable 

 results, but sealing is not necessary if transpiration is not to be meas- 

 ured. Uniformity of temperature, light, and moisture for all plants 

 in the test are absolutely necessary, and it is well to change the relative 

 positions of the baskets frequently, thus balancing any unfavorable 

 conditions due to position. 



It should be borne in mind that this is a method not for a study of 

 the requirements of plants, but for the fertilizer requirements of soils, 

 in which the plants are used as an indicator. It is therefore not nec- 

 essary to grow the plants to maturity ; in fact, it would not be possible 

 to do so successfully in the small amount of soil used. Where differ- 

 ences occur as a result of the fertilizers they manifest themselves 

 almost from the very beginning of plant growth, and it is not necessary 

 or advisable to grow the plants for periods of time exceeding twenty 

 or twenty-five days from the date of planting the seed. 



FRANK D. GARDNER, 

 In Charge of Soil Management. 



Approved : 



JAMES WILSON, 



Secretary of Agriculture. 



MARCH 15, 1905. 



