LXVIII KEPOKT OF THE SECRETARY OF AGRICULTURE. 



pounds per acre 1 foot deep. With these methods it is possi])le to 

 detect throughout the year fertilizers which were applied in the spring, 

 and to trace the movement of these fertilizers from the place where 

 they were applied dow n through the different depths of the soil. It 

 has been found possible to show noticeable differences in the chemical 

 composition of the soil in the same field, in some parts of which the 

 crops are growing well, and in other parts of which they are but 

 poorly developed. It appears that the time has arrived, looked for- 

 ward to with much interest by scientists and practical men alike, when 

 an analysis will show the need of any particular soil for certain fer- 

 tilizers. It is too early to make a positive statement of this kind, but 

 the indications are that this can soon be accomplished. 



A considerable amount of work has been done with these methods 

 on the soils of eastern North Carolina and of Wisconsin, and quite 

 recently many of the important soil types which have been established 

 by the Bureau in Georgia, North Carolina, South Carolina, Virginia, 

 Mar3^1and, New Jersey, Penns3dvania, and Wisconsin have been ana- 

 lyzed by these methods. These types represent all grades of soil, from 

 the most productive to those which are quite unproductive; soils that 

 are adapted to truck crops, tobacco, fruit, wheat, and corn; and soils 

 under a range of climatic conditions with rainfall var3ang from 25 to 

 45 inches per annum. 



The results of this work have led us to look upon the soil moisture 

 as a great nutritive solution existing over the surface of the earth, the 

 composition of which is ever3'whero approximately the same. The 

 soil is a heterogeneous mixture of minerals, the predominant ones 

 being silica, feldspar, mica, and other like silicates, resulting prima- 

 rily from the disintegration and decomposition of igneous rocks, spread 

 out often through the action of water over vast areas of land. All 

 these minerals are but slightly soluble, and it is not unreasonable to 

 expect that such a heterogeneous mixture of silicates in contact with 

 water should 3deld a soil solution having sensibly the same composi- 

 tion and concentration. The older experimenters in Germany found 

 that in making up solutions for water culture or for sand culture the 

 concentration and composition of the nutritive solution must be the 

 same within relatively narrow limits for success in plant development. 

 The results of the past season indicate that the differences in the com- 

 position and concentration of the dissolved material in the soil moisture 

 of various types of soil of widely- different localities and of different 

 agricultural values are little if an3' greater than the differences to be 

 found in one and the same type of soil under good and under poor 

 farm management. 



The "early truck" soil of the Atlantic coast ma3^ be deficient in 

 plant food and may require fertilizers for the best development of the 

 crop, but even with such an application as would make these soils as 



